first commit
This commit is contained in:
admin
89
Seg_All_In_One_MMSeg/configs/ocrnet/README.md
Normal file
89
Seg_All_In_One_MMSeg/configs/ocrnet/README.md
Normal file
@@ -0,0 +1,89 @@
|
||||
# OCRNet
|
||||
|
||||
> [Object-Contextual Representations for Semantic Segmentation](https://arxiv.org/abs/1909.11065)
|
||||
|
||||
## Introduction
|
||||
|
||||
<!-- [ALGORITHM] -->
|
||||
|
||||
<a href="https://github.com/openseg-group/OCNet.pytorch">Official Repo</a>
|
||||
|
||||
<a href="https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86">Code Snippet</a>
|
||||
|
||||
## Abstract
|
||||
|
||||
<!-- [ABSTRACT] -->
|
||||
|
||||
In this paper, we address the problem of semantic segmentation and focus on the context aggregation strategy for robust segmentation. Our motivation is that the label of a pixel is the category of the object that the pixel belongs to. We present a simple yet effective approach, object-contextual representations, characterizing a pixel by exploiting the representation of the corresponding object class. First, we construct object regions based on a feature map supervised by the ground-truth segmentation, and then compute the object region representations. Second, we compute the representation similarity between each pixel and each object region, and augment the representation of each pixel with an object contextual representation, which is a weighted aggregation of all the object region representations according to their similarities with the pixel. We empirically demonstrate that the proposed approach achieves competitive performance on six challenging semantic segmentation benchmarks: Cityscapes, ADE20K, LIP, PASCAL VOC 2012, PASCAL-Context and COCO-Stuff. Notably, we achieved the \\nth{2} place on the Cityscapes leader-board with a single model.
|
||||
|
||||
<!-- [IMAGE] -->
|
||||
|
||||
<div align=center>
|
||||
<img src="https://user-images.githubusercontent.com/24582831/142902197-b06b1e04-57ab-44ac-adc8-cea6695bb236.png" width="70%"/>
|
||||
</div>
|
||||
|
||||
## Results and models
|
||||
|
||||
### Cityscapes
|
||||
|
||||
#### HRNet backbone
|
||||
|
||||
| Method | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) | Device | mIoU | mIoU(ms+flip) | config | download |
|
||||
| ------ | ------------------ | --------- | ------: | -------- | -------------- | ------ | ----: | ------------: | ----------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| OCRNet | HRNetV2p-W18-Small | 512x1024 | 40000 | 3.5 | 10.45 | A100 | 76.61 | 78.01 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024_20230227_145026-6c052a14.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024_20230227_145026.json) |
|
||||
| OCRNet | HRNetV2p-W18 | 512x1024 | 40000 | 4.7 | 7.50 | V100 | 77.72 | 79.49 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18_4xb2-40k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_40k_cityscapes/ocrnet_hr18_512x1024_40k_cityscapes_20200601_033320-401c5bdd.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_40k_cityscapes/ocrnet_hr18_512x1024_40k_cityscapes_20200601_033320.log.json) |
|
||||
| OCRNet | HRNetV2p-W48 | 512x1024 | 40000 | 8 | 4.22 | V100 | 80.58 | 81.79 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr48_4xb2-40k_cityscapes-512x1024.pyy) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_40k_cityscapes/ocrnet_hr48_512x1024_40k_cityscapes_20200601_033336-55b32491.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_40k_cityscapes/ocrnet_hr48_512x1024_40k_cityscapes_20200601_033336.log.json) |
|
||||
| OCRNet | HRNetV2p-W18-Small | 512x1024 | 80000 | - | - | V100 | 77.16 | 78.66 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18s_4xb2-80k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x1024_80k_cityscapes/ocrnet_hr18s_512x1024_80k_cityscapes_20200601_222735-55979e63.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x1024_80k_cityscapes/ocrnet_hr18s_512x1024_80k_cityscapes_20200601_222735.log.json) |
|
||||
| OCRNet | HRNetV2p-W18 | 512x1024 | 80000 | - | - | V100 | 78.57 | 80.46 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18_4xb2-80k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_80k_cityscapes/ocrnet_hr18_512x1024_80k_cityscapes_20200614_230521-c2e1dd4a.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_80k_cityscapes/ocrnet_hr18_512x1024_80k_cityscapes_20200614_230521.log.json) |
|
||||
| OCRNet | HRNetV2p-W48 | 512x1024 | 80000 | - | - | V100 | 80.70 | 81.87 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr48_4xb2-80k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_80k_cityscapes/ocrnet_hr48_512x1024_80k_cityscapes_20200601_222752-9076bcdf.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_80k_cityscapes/ocrnet_hr48_512x1024_80k_cityscapes_20200601_222752.log.json) |
|
||||
| OCRNet | HRNetV2p-W18-Small | 512x1024 | 160000 | - | - | V100 | 78.45 | 79.97 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18s_4xb2-160k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x1024_160k_cityscapes/ocrnet_hr18s_512x1024_160k_cityscapes_20200602_191005-f4a7af28.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x1024_160k_cityscapes/ocrnet_hr18s_512x1024_160k_cityscapes_20200602_191005.log.json) |
|
||||
| OCRNet | HRNetV2p-W18 | 512x1024 | 160000 | - | - | V100 | 79.47 | 80.91 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18_4xb2-160k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_160k_cityscapes/ocrnet_hr18_512x1024_160k_cityscapes_20200602_191001-b9172d0c.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_160k_cityscapes/ocrnet_hr18_512x1024_160k_cityscapes_20200602_191001.log.json) |
|
||||
| OCRNet | HRNetV2p-W48 | 512x1024 | 160000 | - | - | V100 | 81.35 | 82.70 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr48_4xb2-160k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_160k_cityscapes/ocrnet_hr48_512x1024_160k_cityscapes_20200602_191037-dfbf1b0c.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_160k_cityscapes/ocrnet_hr48_512x1024_160k_cityscapes_20200602_191037.log.json) |
|
||||
|
||||
#### ResNet backbone
|
||||
|
||||
| Method | Backbone | Crop Size | Batch Size | Lr schd | Mem (GB) | Inf time (fps) | Device | mIoU | mIoU(ms+flip) | config | download |
|
||||
| ------ | -------- | --------- | ---------- | ------- | -------- | -------------- | ------ | ----- | ------------: | ------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
|
||||
| OCRNet | R-101-D8 | 512x1024 | 8 | 40000 | - | - | V100 | 80.09 | - | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_r101-d8_4xb2-40k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_40k_b8_cityscapes/ocrnet_r101-d8_512x1024_40k_b8_cityscapes_20200717_110721-02ac0f13.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_40k_b8_cityscapes/ocrnet_r101-d8_512x1024_40k_b8_cityscapes_20200717_110721.log.json) |
|
||||
| OCRNet | R-101-D8 | 512x1024 | 16 | 40000 | 8.8 | 3.02 | V100 | 80.30 | - | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_r101-d8_8xb2-40k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_40k_b16_cityscapes/ocrnet_r101-d8_512x1024_40k_b16_cityscapes_20200723_193726-db500f80.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_40k_b16_cityscapes/ocrnet_r101-d8_512x1024_40k_b16_cityscapes_20200723_193726.log.json) |
|
||||
| OCRNet | R-101-D8 | 512x1024 | 16 | 80000 | 8.8 | 3.02 | V100 | 80.81 | - | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_r101-d8_8xb2-80k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_80k_b16_cityscapes/ocrnet_r101-d8_512x1024_80k_b16_cityscapes_20200723_192421-78688424.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_80k_b16_cityscapes/ocrnet_r101-d8_512x1024_80k_b16_cityscapes_20200723_192421.log.json) |
|
||||
|
||||
### ADE20K
|
||||
|
||||
| Method | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) | Device | mIoU | mIoU(ms+flip) | config | download |
|
||||
| ------ | ------------------ | --------- | ------: | -------- | -------------- | ------ | ----: | ------------: | ----------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| OCRNet | HRNetV2p-W18-Small | 512x512 | 80000 | 6.7 | 28.98 | V100 | 35.06 | 35.80 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18s_4xb4-80k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_80k_ade20k/ocrnet_hr18s_512x512_80k_ade20k_20200615_055600-e80b62af.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_80k_ade20k/ocrnet_hr18s_512x512_80k_ade20k_20200615_055600.log.json) |
|
||||
| OCRNet | HRNetV2p-W18 | 512x512 | 80000 | 7.9 | 18.93 | V100 | 37.79 | 39.16 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18_4xb4-80k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_80k_ade20k/ocrnet_hr18_512x512_80k_ade20k_20200615_053157-d173d83b.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_80k_ade20k/ocrnet_hr18_512x512_80k_ade20k_20200615_053157.log.json) |
|
||||
| OCRNet | HRNetV2p-W48 | 512x512 | 80000 | 11.2 | 16.99 | V100 | 43.00 | 44.30 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr48_4xb4-80k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_80k_ade20k/ocrnet_hr48_512x512_80k_ade20k_20200615_021518-d168c2d1.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_80k_ade20k/ocrnet_hr48_512x512_80k_ade20k_20200615_021518.log.json) |
|
||||
| OCRNet | HRNetV2p-W18-Small | 512x512 | 160000 | - | - | V100 | 37.19 | 38.40 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18s_4xb4-80k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_160k_ade20k/ocrnet_hr18s_512x512_160k_ade20k_20200615_184505-8e913058.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_160k_ade20k/ocrnet_hr18s_512x512_160k_ade20k_20200615_184505.log.json) |
|
||||
| OCRNet | HRNetV2p-W18 | 512x512 | 160000 | - | - | V100 | 39.32 | 40.80 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18_4xb4-80k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_160k_ade20k/ocrnet_hr18_512x512_160k_ade20k_20200615_200940-d8fcd9d1.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_160k_ade20k/ocrnet_hr18_512x512_160k_ade20k_20200615_200940.log.json) |
|
||||
| OCRNet | HRNetV2p-W48 | 512x512 | 160000 | - | - | V100 | 43.25 | 44.88 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr48_4xb4-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_160k_ade20k/ocrnet_hr48_512x512_160k_ade20k_20200615_184705-a073726d.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_160k_ade20k/ocrnet_hr48_512x512_160k_ade20k_20200615_184705.log.json) |
|
||||
|
||||
### Pascal VOC 2012 + Aug
|
||||
|
||||
| Method | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) | Device | mIoU | mIoU(ms+flip) | config | download |
|
||||
| ------ | ------------------ | --------- | ------: | -------- | -------------- | ------ | ----: | ------------: | ------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
|
||||
| OCRNet | HRNetV2p-W18-Small | 512x512 | 20000 | 3.5 | 31.55 | V100 | 71.70 | 73.84 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18s_4xb4-20k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_20k_voc12aug/ocrnet_hr18s_512x512_20k_voc12aug_20200617_233913-02b04fcb.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_20k_voc12aug/ocrnet_hr18s_512x512_20k_voc12aug_20200617_233913.log.json) |
|
||||
| OCRNet | HRNetV2p-W18 | 512x512 | 20000 | 4.7 | 19.91 | V100 | 74.75 | 77.11 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18_4xb4-20k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_20k_voc12aug/ocrnet_hr18_512x512_20k_voc12aug_20200617_233932-8954cbb7.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_20k_voc12aug/ocrnet_hr18_512x512_20k_voc12aug_20200617_233932.log.json) |
|
||||
| OCRNet | HRNetV2p-W48 | 512x512 | 20000 | 8.1 | 17.83 | V100 | 77.72 | 79.87 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr48_4xb4-20k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_20k_voc12aug/ocrnet_hr48_512x512_20k_voc12aug_20200617_233932-9e82080a.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_20k_voc12aug/ocrnet_hr48_512x512_20k_voc12aug_20200617_233932.log.json) |
|
||||
| OCRNet | HRNetV2p-W18-Small | 512x512 | 40000 | - | - | V100 | 72.76 | 74.60 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18s_4xb4-40k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_40k_voc12aug/ocrnet_hr18s_512x512_40k_voc12aug_20200614_002025-42b587ac.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_40k_voc12aug/ocrnet_hr18s_512x512_40k_voc12aug_20200614_002025.log.json) |
|
||||
| OCRNet | HRNetV2p-W18 | 512x512 | 40000 | - | - | V100 | 74.98 | 77.40 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr18_4xb4-40k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_40k_voc12aug/ocrnet_hr18_512x512_40k_voc12aug_20200614_015958-714302be.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_40k_voc12aug/ocrnet_hr18_512x512_40k_voc12aug_20200614_015958.log.json) |
|
||||
| OCRNet | HRNetV2p-W48 | 512x512 | 40000 | - | - | V100 | 77.14 | 79.71 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/ocrnet/ocrnet_hr48_4xb4-40k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_40k_voc12aug/ocrnet_hr48_512x512_40k_voc12aug_20200614_015958-255bc5ce.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_40k_voc12aug/ocrnet_hr48_512x512_40k_voc12aug_20200614_015958.log.json) |
|
||||
|
||||
## Citation
|
||||
|
||||
```bibtex
|
||||
@article{YuanW18,
|
||||
title={Ocnet: Object context network for scene parsing},
|
||||
author={Yuhui Yuan and Jingdong Wang},
|
||||
booktitle={arXiv preprint arXiv:1809.00916},
|
||||
year={2018}
|
||||
}
|
||||
|
||||
@article{YuanCW20,
|
||||
title={Object-Contextual Representations for Semantic Segmentation},
|
||||
author={Yuhui Yuan and Xilin Chen and Jingdong Wang},
|
||||
booktitle={ECCV},
|
||||
year={2020}
|
||||
}
|
||||
```
|
||||
577
Seg_All_In_One_MMSeg/configs/ocrnet/metafile.yaml
Normal file
577
Seg_All_In_One_MMSeg/configs/ocrnet/metafile.yaml
Normal file
@@ -0,0 +1,577 @@
|
||||
Collections:
|
||||
- Name: OCRNet
|
||||
License: Apache License 2.0
|
||||
Metadata:
|
||||
Training Data:
|
||||
- Cityscapes
|
||||
- '# HRNet backbone'
|
||||
- '# ResNet backbone'
|
||||
- ADE20K
|
||||
- Pascal VOC 2012 + Aug
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
README: configs/ocrnet/README.md
|
||||
Frameworks:
|
||||
- PyTorch
|
||||
Models:
|
||||
- Name: ocrnet_hr18s_4xb2-40k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 76.61
|
||||
mIoU(ms+flip): 78.01
|
||||
Config: configs/ocrnet/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W18-Small
|
||||
- OCRNet
|
||||
Training Resources: 4x A100 GPUS
|
||||
Memory (GB): 3.5
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024_20230227_145026-6c052a14.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024/ocrnet_hr18s_4xb2-40k_cityscapes-512x1024_20230227_145026.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18_4xb2-40k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 77.72
|
||||
mIoU(ms+flip): 79.49
|
||||
Config: configs/ocrnet/ocrnet_hr18_4xb2-40k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W18
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Memory (GB): 4.7
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_40k_cityscapes/ocrnet_hr18_512x1024_40k_cityscapes_20200601_033320-401c5bdd.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_40k_cityscapes/ocrnet_hr18_512x1024_40k_cityscapes_20200601_033320.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr48_4xb2-40k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 80.58
|
||||
mIoU(ms+flip): 81.79
|
||||
Config: configs/ocrnet/ocrnet_hr48_4xb2-40k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W48
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Memory (GB): 8.0
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_40k_cityscapes/ocrnet_hr48_512x1024_40k_cityscapes_20200601_033336-55b32491.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_40k_cityscapes/ocrnet_hr48_512x1024_40k_cityscapes_20200601_033336.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18s_4xb2-80k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 77.16
|
||||
mIoU(ms+flip): 78.66
|
||||
Config: configs/ocrnet/ocrnet_hr18s_4xb2-80k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W18-Small
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x1024_80k_cityscapes/ocrnet_hr18s_512x1024_80k_cityscapes_20200601_222735-55979e63.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x1024_80k_cityscapes/ocrnet_hr18s_512x1024_80k_cityscapes_20200601_222735.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18_4xb2-80k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 78.57
|
||||
mIoU(ms+flip): 80.46
|
||||
Config: configs/ocrnet/ocrnet_hr18_4xb2-80k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W18
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_80k_cityscapes/ocrnet_hr18_512x1024_80k_cityscapes_20200614_230521-c2e1dd4a.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_80k_cityscapes/ocrnet_hr18_512x1024_80k_cityscapes_20200614_230521.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr48_4xb2-80k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 80.7
|
||||
mIoU(ms+flip): 81.87
|
||||
Config: configs/ocrnet/ocrnet_hr48_4xb2-80k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W48
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_80k_cityscapes/ocrnet_hr48_512x1024_80k_cityscapes_20200601_222752-9076bcdf.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_80k_cityscapes/ocrnet_hr48_512x1024_80k_cityscapes_20200601_222752.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18s_4xb2-160k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 78.45
|
||||
mIoU(ms+flip): 79.97
|
||||
Config: configs/ocrnet/ocrnet_hr18s_4xb2-160k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W18-Small
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x1024_160k_cityscapes/ocrnet_hr18s_512x1024_160k_cityscapes_20200602_191005-f4a7af28.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x1024_160k_cityscapes/ocrnet_hr18s_512x1024_160k_cityscapes_20200602_191005.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18_4xb2-160k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 79.47
|
||||
mIoU(ms+flip): 80.91
|
||||
Config: configs/ocrnet/ocrnet_hr18_4xb2-160k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W18
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_160k_cityscapes/ocrnet_hr18_512x1024_160k_cityscapes_20200602_191001-b9172d0c.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x1024_160k_cityscapes/ocrnet_hr18_512x1024_160k_cityscapes_20200602_191001.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr48_4xb2-160k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# HRNet backbone'
|
||||
Metrics:
|
||||
mIoU: 81.35
|
||||
mIoU(ms+flip): 82.7
|
||||
Config: configs/ocrnet/ocrnet_hr48_4xb2-160k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# HRNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- HRNetV2p-W48
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_160k_cityscapes/ocrnet_hr48_512x1024_160k_cityscapes_20200602_191037-dfbf1b0c.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x1024_160k_cityscapes/ocrnet_hr48_512x1024_160k_cityscapes_20200602_191037.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_r101-d8_4xb2-40k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# ResNet backbone'
|
||||
Metrics:
|
||||
mIoU: 80.09
|
||||
Config: configs/ocrnet/ocrnet_r101-d8_4xb2-40k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# ResNet backbone'
|
||||
Batch Size: 8
|
||||
Architecture:
|
||||
- R-101-D8
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_40k_b8_cityscapes/ocrnet_r101-d8_512x1024_40k_b8_cityscapes_20200717_110721-02ac0f13.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_40k_b8_cityscapes/ocrnet_r101-d8_512x1024_40k_b8_cityscapes_20200717_110721.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_r101-d8_8xb2-40k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# ResNet backbone'
|
||||
Metrics:
|
||||
mIoU: 80.3
|
||||
Config: configs/ocrnet/ocrnet_r101-d8_8xb2-40k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# ResNet backbone'
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- R-101-D8
|
||||
- OCRNet
|
||||
Training Resources: 8x V100 GPUS
|
||||
Memory (GB): 8.8
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_40k_b16_cityscapes/ocrnet_r101-d8_512x1024_40k_b16_cityscapes_20200723_193726-db500f80.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_40k_b16_cityscapes/ocrnet_r101-d8_512x1024_40k_b16_cityscapes_20200723_193726.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_r101-d8_8xb2-80k_cityscapes-512x1024
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: '# ResNet backbone'
|
||||
Metrics:
|
||||
mIoU: 80.81
|
||||
Config: configs/ocrnet/ocrnet_r101-d8_8xb2-80k_cityscapes-512x1024.py
|
||||
Metadata:
|
||||
Training Data: '# ResNet backbone'
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- R-101-D8
|
||||
- OCRNet
|
||||
Training Resources: 8x V100 GPUS
|
||||
Memory (GB): 8.8
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_80k_b16_cityscapes/ocrnet_r101-d8_512x1024_80k_b16_cityscapes_20200723_192421-78688424.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_r101-d8_512x1024_80k_b16_cityscapes/ocrnet_r101-d8_512x1024_80k_b16_cityscapes_20200723_192421.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18s_4xb4-80k_ade20k-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: ADE20K
|
||||
Metrics:
|
||||
mIoU: 35.06
|
||||
mIoU(ms+flip): 35.8
|
||||
Config: configs/ocrnet/ocrnet_hr18s_4xb4-80k_ade20k-512x512.py
|
||||
Metadata:
|
||||
Training Data: ADE20K
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W18-Small
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Memory (GB): 6.7
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_80k_ade20k/ocrnet_hr18s_512x512_80k_ade20k_20200615_055600-e80b62af.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_80k_ade20k/ocrnet_hr18s_512x512_80k_ade20k_20200615_055600.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18_4xb4-80k_ade20k-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: ADE20K
|
||||
Metrics:
|
||||
mIoU: 37.79
|
||||
mIoU(ms+flip): 39.16
|
||||
Config: configs/ocrnet/ocrnet_hr18_4xb4-80k_ade20k-512x512.py
|
||||
Metadata:
|
||||
Training Data: ADE20K
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W18
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Memory (GB): 7.9
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_80k_ade20k/ocrnet_hr18_512x512_80k_ade20k_20200615_053157-d173d83b.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_80k_ade20k/ocrnet_hr18_512x512_80k_ade20k_20200615_053157.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr48_4xb4-80k_ade20k-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: ADE20K
|
||||
Metrics:
|
||||
mIoU: 43.0
|
||||
mIoU(ms+flip): 44.3
|
||||
Config: configs/ocrnet/ocrnet_hr48_4xb4-80k_ade20k-512x512.py
|
||||
Metadata:
|
||||
Training Data: ADE20K
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W48
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Memory (GB): 11.2
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_80k_ade20k/ocrnet_hr48_512x512_80k_ade20k_20200615_021518-d168c2d1.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_80k_ade20k/ocrnet_hr48_512x512_80k_ade20k_20200615_021518.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18s_4xb4-80k_ade20k-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: ADE20K
|
||||
Metrics:
|
||||
mIoU: 37.19
|
||||
mIoU(ms+flip): 38.4
|
||||
Config: configs/ocrnet/ocrnet_hr18s_4xb4-80k_ade20k-512x512.py
|
||||
Metadata:
|
||||
Training Data: ADE20K
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W18-Small
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_160k_ade20k/ocrnet_hr18s_512x512_160k_ade20k_20200615_184505-8e913058.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_160k_ade20k/ocrnet_hr18s_512x512_160k_ade20k_20200615_184505.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18_4xb4-80k_ade20k-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: ADE20K
|
||||
Metrics:
|
||||
mIoU: 39.32
|
||||
mIoU(ms+flip): 40.8
|
||||
Config: configs/ocrnet/ocrnet_hr18_4xb4-80k_ade20k-512x512.py
|
||||
Metadata:
|
||||
Training Data: ADE20K
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W18
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_160k_ade20k/ocrnet_hr18_512x512_160k_ade20k_20200615_200940-d8fcd9d1.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_160k_ade20k/ocrnet_hr18_512x512_160k_ade20k_20200615_200940.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr48_4xb4-160k_ade20k-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: ADE20K
|
||||
Metrics:
|
||||
mIoU: 43.25
|
||||
mIoU(ms+flip): 44.88
|
||||
Config: configs/ocrnet/ocrnet_hr48_4xb4-160k_ade20k-512x512.py
|
||||
Metadata:
|
||||
Training Data: ADE20K
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W48
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_160k_ade20k/ocrnet_hr48_512x512_160k_ade20k_20200615_184705-a073726d.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_160k_ade20k/ocrnet_hr48_512x512_160k_ade20k_20200615_184705.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18s_4xb4-20k_voc12aug-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: Pascal VOC 2012 + Aug
|
||||
Metrics:
|
||||
mIoU: 71.7
|
||||
mIoU(ms+flip): 73.84
|
||||
Config: configs/ocrnet/ocrnet_hr18s_4xb4-20k_voc12aug-512x512.py
|
||||
Metadata:
|
||||
Training Data: Pascal VOC 2012 + Aug
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W18-Small
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Memory (GB): 3.5
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_20k_voc12aug/ocrnet_hr18s_512x512_20k_voc12aug_20200617_233913-02b04fcb.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_20k_voc12aug/ocrnet_hr18s_512x512_20k_voc12aug_20200617_233913.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18_4xb4-20k_voc12aug-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: Pascal VOC 2012 + Aug
|
||||
Metrics:
|
||||
mIoU: 74.75
|
||||
mIoU(ms+flip): 77.11
|
||||
Config: configs/ocrnet/ocrnet_hr18_4xb4-20k_voc12aug-512x512.py
|
||||
Metadata:
|
||||
Training Data: Pascal VOC 2012 + Aug
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W18
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Memory (GB): 4.7
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_20k_voc12aug/ocrnet_hr18_512x512_20k_voc12aug_20200617_233932-8954cbb7.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_20k_voc12aug/ocrnet_hr18_512x512_20k_voc12aug_20200617_233932.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr48_4xb4-20k_voc12aug-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: Pascal VOC 2012 + Aug
|
||||
Metrics:
|
||||
mIoU: 77.72
|
||||
mIoU(ms+flip): 79.87
|
||||
Config: configs/ocrnet/ocrnet_hr48_4xb4-20k_voc12aug-512x512.py
|
||||
Metadata:
|
||||
Training Data: Pascal VOC 2012 + Aug
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W48
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Memory (GB): 8.1
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_20k_voc12aug/ocrnet_hr48_512x512_20k_voc12aug_20200617_233932-9e82080a.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_20k_voc12aug/ocrnet_hr48_512x512_20k_voc12aug_20200617_233932.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18s_4xb4-40k_voc12aug-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: Pascal VOC 2012 + Aug
|
||||
Metrics:
|
||||
mIoU: 72.76
|
||||
mIoU(ms+flip): 74.6
|
||||
Config: configs/ocrnet/ocrnet_hr18s_4xb4-40k_voc12aug-512x512.py
|
||||
Metadata:
|
||||
Training Data: Pascal VOC 2012 + Aug
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W18-Small
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_40k_voc12aug/ocrnet_hr18s_512x512_40k_voc12aug_20200614_002025-42b587ac.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18s_512x512_40k_voc12aug/ocrnet_hr18s_512x512_40k_voc12aug_20200614_002025.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr18_4xb4-40k_voc12aug-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: Pascal VOC 2012 + Aug
|
||||
Metrics:
|
||||
mIoU: 74.98
|
||||
mIoU(ms+flip): 77.4
|
||||
Config: configs/ocrnet/ocrnet_hr18_4xb4-40k_voc12aug-512x512.py
|
||||
Metadata:
|
||||
Training Data: Pascal VOC 2012 + Aug
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W18
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_40k_voc12aug/ocrnet_hr18_512x512_40k_voc12aug_20200614_015958-714302be.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr18_512x512_40k_voc12aug/ocrnet_hr18_512x512_40k_voc12aug_20200614_015958.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
- Name: ocrnet_hr48_4xb4-40k_voc12aug-512x512
|
||||
In Collection: OCRNet
|
||||
Results:
|
||||
Task: Semantic Segmentation
|
||||
Dataset: Pascal VOC 2012 + Aug
|
||||
Metrics:
|
||||
mIoU: 77.14
|
||||
mIoU(ms+flip): 79.71
|
||||
Config: configs/ocrnet/ocrnet_hr48_4xb4-40k_voc12aug-512x512.py
|
||||
Metadata:
|
||||
Training Data: Pascal VOC 2012 + Aug
|
||||
Batch Size: 16
|
||||
Architecture:
|
||||
- HRNetV2p-W48
|
||||
- OCRNet
|
||||
Training Resources: 4x V100 GPUS
|
||||
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_40k_voc12aug/ocrnet_hr48_512x512_40k_voc12aug_20200614_015958-255bc5ce.pth
|
||||
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/ocrnet/ocrnet_hr48_512x512_40k_voc12aug/ocrnet_hr48_512x512_40k_voc12aug_20200614_015958.log.json
|
||||
Paper:
|
||||
Title: Object-Contextual Representations for Semantic Segmentation
|
||||
URL: https://arxiv.org/abs/1909.11065
|
||||
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/ocr_head.py#L86
|
||||
Framework: PyTorch
|
||||
@@ -0,0 +1,7 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_hr18.py', '../_base_/datasets/cityscapes.py',
|
||||
'../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py'
|
||||
]
|
||||
crop_size = (512, 1024)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
model = dict(data_preprocessor=data_preprocessor)
|
||||
@@ -0,0 +1,7 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_hr18.py', '../_base_/datasets/cityscapes.py',
|
||||
'../_base_/default_runtime.py', '../_base_/schedules/schedule_40k.py'
|
||||
]
|
||||
crop_size = (512, 1024)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
model = dict(data_preprocessor=data_preprocessor)
|
||||
@@ -0,0 +1,7 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_hr18.py', '../_base_/datasets/cityscapes.py',
|
||||
'../_base_/default_runtime.py', '../_base_/schedules/schedule_80k.py'
|
||||
]
|
||||
crop_size = (512, 1024)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
model = dict(data_preprocessor=data_preprocessor)
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_hr18.py', '../_base_/datasets/ade20k.py',
|
||||
'../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py'
|
||||
]
|
||||
crop_size = (512, 512)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
data_preprocessor=data_preprocessor,
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[18, 36, 72, 144],
|
||||
channels=sum([18, 36, 72, 144]),
|
||||
in_index=(0, 1, 2, 3),
|
||||
input_transform='resize_concat',
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=150,
|
||||
norm_cfg=norm_cfg,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[18, 36, 72, 144],
|
||||
in_index=(0, 1, 2, 3),
|
||||
input_transform='resize_concat',
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
dropout_ratio=-1,
|
||||
num_classes=150,
|
||||
norm_cfg=norm_cfg,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
|
||||
])
|
||||
@@ -0,0 +1,40 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_hr18.py',
|
||||
'../_base_/datasets/pascal_voc12_aug.py', '../_base_/default_runtime.py',
|
||||
'../_base_/schedules/schedule_20k.py'
|
||||
]
|
||||
crop_size = (512, 512)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
data_preprocessor=data_preprocessor,
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[18, 36, 72, 144],
|
||||
channels=sum([18, 36, 72, 144]),
|
||||
in_index=(0, 1, 2, 3),
|
||||
input_transform='resize_concat',
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=21,
|
||||
norm_cfg=norm_cfg,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[18, 36, 72, 144],
|
||||
in_index=(0, 1, 2, 3),
|
||||
input_transform='resize_concat',
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
dropout_ratio=-1,
|
||||
num_classes=21,
|
||||
norm_cfg=norm_cfg,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
|
||||
])
|
||||
@@ -0,0 +1,40 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_hr18.py',
|
||||
'../_base_/datasets/pascal_voc12_aug.py', '../_base_/default_runtime.py',
|
||||
'../_base_/schedules/schedule_40k.py'
|
||||
]
|
||||
crop_size = (512, 512)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
data_preprocessor=data_preprocessor,
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[18, 36, 72, 144],
|
||||
channels=sum([18, 36, 72, 144]),
|
||||
in_index=(0, 1, 2, 3),
|
||||
input_transform='resize_concat',
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=21,
|
||||
norm_cfg=norm_cfg,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[18, 36, 72, 144],
|
||||
in_index=(0, 1, 2, 3),
|
||||
input_transform='resize_concat',
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
dropout_ratio=-1,
|
||||
num_classes=21,
|
||||
norm_cfg=norm_cfg,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
|
||||
])
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_hr18.py', '../_base_/datasets/ade20k.py',
|
||||
'../_base_/default_runtime.py', '../_base_/schedules/schedule_80k.py'
|
||||
]
|
||||
crop_size = (512, 512)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
data_preprocessor=data_preprocessor,
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[18, 36, 72, 144],
|
||||
channels=sum([18, 36, 72, 144]),
|
||||
in_index=(0, 1, 2, 3),
|
||||
input_transform='resize_concat',
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=150,
|
||||
norm_cfg=norm_cfg,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[18, 36, 72, 144],
|
||||
in_index=(0, 1, 2, 3),
|
||||
input_transform='resize_concat',
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
dropout_ratio=-1,
|
||||
num_classes=150,
|
||||
norm_cfg=norm_cfg,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0)),
|
||||
])
|
||||
@@ -0,0 +1,9 @@
|
||||
_base_ = './ocrnet_hr18_4xb2-160k_cityscapes-512x1024.py'
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w18_small',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage1=dict(num_blocks=(2, )),
|
||||
stage2=dict(num_blocks=(2, 2)),
|
||||
stage3=dict(num_modules=3, num_blocks=(2, 2, 2)),
|
||||
stage4=dict(num_modules=2, num_blocks=(2, 2, 2, 2)))))
|
||||
@@ -0,0 +1,9 @@
|
||||
_base_ = './ocrnet_hr18_4xb2-40k_cityscapes-512x1024.py'
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w18_small',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage1=dict(num_blocks=(2, )),
|
||||
stage2=dict(num_blocks=(2, 2)),
|
||||
stage3=dict(num_modules=3, num_blocks=(2, 2, 2)),
|
||||
stage4=dict(num_modules=2, num_blocks=(2, 2, 2, 2)))))
|
||||
@@ -0,0 +1,9 @@
|
||||
_base_ = './ocrnet_hr18_4xb2-80k_cityscapes-512x1024.py'
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w18_small',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage1=dict(num_blocks=(2, )),
|
||||
stage2=dict(num_blocks=(2, 2)),
|
||||
stage3=dict(num_modules=3, num_blocks=(2, 2, 2)),
|
||||
stage4=dict(num_modules=2, num_blocks=(2, 2, 2, 2)))))
|
||||
@@ -0,0 +1,9 @@
|
||||
_base_ = './ocrnet_hr18_4xb4-160k_ade20k-512x512.py'
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w18_small',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage1=dict(num_blocks=(2, )),
|
||||
stage2=dict(num_blocks=(2, 2)),
|
||||
stage3=dict(num_modules=3, num_blocks=(2, 2, 2)),
|
||||
stage4=dict(num_modules=2, num_blocks=(2, 2, 2, 2)))))
|
||||
@@ -0,0 +1,9 @@
|
||||
_base_ = './ocrnet_hr18_4xb4-20k_voc12aug-512x512.py'
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w18_small',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage1=dict(num_blocks=(2, )),
|
||||
stage2=dict(num_blocks=(2, 2)),
|
||||
stage3=dict(num_modules=3, num_blocks=(2, 2, 2)),
|
||||
stage4=dict(num_modules=2, num_blocks=(2, 2, 2, 2)))))
|
||||
@@ -0,0 +1,9 @@
|
||||
_base_ = './ocrnet_hr18_4xb4-40k_voc12aug-512x512.py'
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w18_small',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage1=dict(num_blocks=(2, )),
|
||||
stage2=dict(num_blocks=(2, 2)),
|
||||
stage3=dict(num_modules=3, num_blocks=(2, 2, 2)),
|
||||
stage4=dict(num_modules=2, num_blocks=(2, 2, 2, 2)))))
|
||||
@@ -0,0 +1,9 @@
|
||||
_base_ = './ocrnet_hr18_4xb4-80k_ade20k-512x512.py'
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w18_small',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage1=dict(num_blocks=(2, )),
|
||||
stage2=dict(num_blocks=(2, 2)),
|
||||
stage3=dict(num_modules=3, num_blocks=(2, 2, 2)),
|
||||
stage4=dict(num_modules=2, num_blocks=(2, 2, 2, 2)))))
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = './ocrnet_hr18_4xb2-160k_cityscapes-512x1024.py'
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w48',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage2=dict(num_channels=(48, 96)),
|
||||
stage3=dict(num_channels=(48, 96, 192)),
|
||||
stage4=dict(num_channels=(48, 96, 192, 384)))),
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=sum([48, 96, 192, 384]),
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
norm_cfg=norm_cfg,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=19,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
norm_cfg=norm_cfg,
|
||||
dropout_ratio=-1,
|
||||
num_classes=19,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0))
|
||||
])
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = './ocrnet_hr18_4xb2-40k_cityscapes-512x1024.py'
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w48',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage2=dict(num_channels=(48, 96)),
|
||||
stage3=dict(num_channels=(48, 96, 192)),
|
||||
stage4=dict(num_channels=(48, 96, 192, 384)))),
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=sum([48, 96, 192, 384]),
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
norm_cfg=norm_cfg,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=19,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
norm_cfg=norm_cfg,
|
||||
dropout_ratio=-1,
|
||||
num_classes=19,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0))
|
||||
])
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = './ocrnet_hr18_4xb2-80k_cityscapes-512x1024.py'
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w48',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage2=dict(num_channels=(48, 96)),
|
||||
stage3=dict(num_channels=(48, 96, 192)),
|
||||
stage4=dict(num_channels=(48, 96, 192, 384)))),
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=sum([48, 96, 192, 384]),
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
norm_cfg=norm_cfg,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=19,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
norm_cfg=norm_cfg,
|
||||
dropout_ratio=-1,
|
||||
num_classes=19,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0))
|
||||
])
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = './ocrnet_hr18_4xb4-160k_ade20k-512x512.py'
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w48',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage2=dict(num_channels=(48, 96)),
|
||||
stage3=dict(num_channels=(48, 96, 192)),
|
||||
stage4=dict(num_channels=(48, 96, 192, 384)))),
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=sum([48, 96, 192, 384]),
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
norm_cfg=norm_cfg,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=150,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
norm_cfg=norm_cfg,
|
||||
dropout_ratio=-1,
|
||||
num_classes=150,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0))
|
||||
])
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = './ocrnet_hr18_4xb4-20k_voc12aug-512x512.py'
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w48',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage2=dict(num_channels=(48, 96)),
|
||||
stage3=dict(num_channels=(48, 96, 192)),
|
||||
stage4=dict(num_channels=(48, 96, 192, 384)))),
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=sum([48, 96, 192, 384]),
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
norm_cfg=norm_cfg,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=21,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
norm_cfg=norm_cfg,
|
||||
dropout_ratio=-1,
|
||||
num_classes=21,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0))
|
||||
])
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = './ocrnet_hr18_4xb4-40k_voc12aug-512x512.py'
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w48',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage2=dict(num_channels=(48, 96)),
|
||||
stage3=dict(num_channels=(48, 96, 192)),
|
||||
stage4=dict(num_channels=(48, 96, 192, 384)))),
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=sum([48, 96, 192, 384]),
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
norm_cfg=norm_cfg,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=21,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
norm_cfg=norm_cfg,
|
||||
dropout_ratio=-1,
|
||||
num_classes=21,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0))
|
||||
])
|
||||
@@ -0,0 +1,39 @@
|
||||
_base_ = './ocrnet_hr18_4xb4-80k_ade20k-512x512.py'
|
||||
norm_cfg = dict(type='SyncBN', requires_grad=True)
|
||||
model = dict(
|
||||
pretrained='open-mmlab://msra/hrnetv2_w48',
|
||||
backbone=dict(
|
||||
extra=dict(
|
||||
stage2=dict(num_channels=(48, 96)),
|
||||
stage3=dict(num_channels=(48, 96, 192)),
|
||||
stage4=dict(num_channels=(48, 96, 192, 384)))),
|
||||
decode_head=[
|
||||
dict(
|
||||
type='FCNHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=sum([48, 96, 192, 384]),
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
kernel_size=1,
|
||||
num_convs=1,
|
||||
norm_cfg=norm_cfg,
|
||||
concat_input=False,
|
||||
dropout_ratio=-1,
|
||||
num_classes=150,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=0.4)),
|
||||
dict(
|
||||
type='OCRHead',
|
||||
in_channels=[48, 96, 192, 384],
|
||||
channels=512,
|
||||
ocr_channels=256,
|
||||
input_transform='resize_concat',
|
||||
in_index=(0, 1, 2, 3),
|
||||
norm_cfg=norm_cfg,
|
||||
dropout_ratio=-1,
|
||||
num_classes=150,
|
||||
align_corners=False,
|
||||
loss_decode=dict(
|
||||
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0))
|
||||
])
|
||||
@@ -0,0 +1,10 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_r50-d8.py', '../_base_/datasets/cityscapes.py',
|
||||
'../_base_/default_runtime.py', '../_base_/schedules/schedule_40k.py'
|
||||
]
|
||||
crop_size = (512, 1024)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
model = dict(
|
||||
data_preprocessor=data_preprocessor,
|
||||
pretrained='open-mmlab://resnet101_v1c',
|
||||
backbone=dict(depth=101))
|
||||
@@ -0,0 +1,21 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_r50-d8.py', '../_base_/datasets/cityscapes.py',
|
||||
'../_base_/default_runtime.py', '../_base_/schedules/schedule_40k.py'
|
||||
]
|
||||
crop_size = (512, 1024)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
model = dict(
|
||||
data_preprocessor=data_preprocessor,
|
||||
pretrained='open-mmlab://resnet101_v1c',
|
||||
backbone=dict(depth=101))
|
||||
optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0005)
|
||||
optim_wrapper = dict(type='OptimWrapper', optimizer=optimizer)
|
||||
param_scheduler = [
|
||||
dict(
|
||||
type='PolyLR',
|
||||
eta_min=2e-4,
|
||||
power=0.9,
|
||||
begin=0,
|
||||
end=40000,
|
||||
by_epoch=False)
|
||||
]
|
||||
@@ -0,0 +1,21 @@
|
||||
_base_ = [
|
||||
'../_base_/models/ocrnet_r50-d8.py', '../_base_/datasets/cityscapes.py',
|
||||
'../_base_/default_runtime.py', '../_base_/schedules/schedule_80k.py'
|
||||
]
|
||||
crop_size = (512, 1024)
|
||||
data_preprocessor = dict(size=crop_size)
|
||||
model = dict(
|
||||
data_preprocessor=data_preprocessor,
|
||||
pretrained='open-mmlab://resnet101_v1c',
|
||||
backbone=dict(depth=101))
|
||||
optimizer = dict(type='SGD', lr=0.02, momentum=0.9, weight_decay=0.0005)
|
||||
optim_wrapper = dict(type='OptimWrapper', optimizer=optimizer)
|
||||
param_scheduler = [
|
||||
dict(
|
||||
type='PolyLR',
|
||||
eta_min=2e-4,
|
||||
power=0.9,
|
||||
begin=0,
|
||||
end=40000,
|
||||
by_epoch=False)
|
||||
]
|
||||
Reference in New Issue
Block a user