Skip to content

GitLab

Commit b1238b9d authored by zhe chen's avatar zhe chen
Browse files

Update README.md

parent f250aa41
Hide whitespace changes
Inline Side-by-side
Showing with 23 additions and 6 deletions
segmentation/README.md
View file @ b1238b9d
......@@ -138,9 +138,9 @@ Prepare datasets according to the [guidelines](https://github.com/open-mmlab/mms
<br>
<div>
| method | backbone | resolution | mIoU (ss) | #params | FLOPs | Config | Download |
| :---------: | :-----------: | :--------: | :-------: | :-----: | :---: | :-----------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| Mask2Former | InternImage-H | 896x896 | 52.6 | 1.31B | 4635G | [config](./configs/coco_stuff164k/mask2former_internimage_h_896_80k_cocostuff164k.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/mask2former_internimage_h_896_80k_cocostuff164k.pth) \| [log](https://huggingface.co/OpenGVLab/InternImage/raw/main/mask2former_internimage_h_896_80k_cocostuff164k.log.json) |
| method | backbone | resolution | mIoU (ss/ms) | #params | FLOPs | Config | Download |
| :---------: | :-----------: | :--------: | :----------: | :-----: | :---: | :-----------------------------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| Mask2Former | InternImage-H | 896x896 | 52.6 / 52.8 | 1.31B | 4635G | [config](./configs/coco_stuff164k/mask2former_internimage_h_896_80k_cocostuff164k.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/mask2former_internimage_h_896_80k_cocostuff164k.pth) \| [log](https://huggingface.co/OpenGVLab/InternImage/raw/main/mask2former_internimage_h_896_80k_cocostuff164k.log.json) |
</div>
......@@ -185,6 +185,23 @@ Prepare datasets according to the [guidelines](https://github.com/open-mmlab/mms
</details>
<details>
<summary> Dataset: Mapillary </summary>
<br>
<div>
| method | backbone | resolution | #param | FLOPs | Config | Download |
| :---------: | :------------: | :--------: | :----: | :---: | :------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------------------: |
| UperNet | InternImage-L | 512x1024 | 256M | 3234G | [config](./configs/mapillary/upernet_internimage_l_512x1024_80k_mapillary.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/upernet_internimage_l_512x1024_80k_mapillary.pth) |
| UperNet | InternImage-XL | 512x1024 | 368M | 4022G | [config](./configs/mapillary/upernet_internimage_xl_512x1024_80k_mapillary.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/upernet_internimage_xl_512x1024_80k_mapillary.pth) |
| SegFormer | InternImage-L | 512x1024 | 220M | 1580G | [config](./configs/mapillary/segformer_internimage_l_512x1024_80k_mapillary.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/segformer_internimage_l_512x1024_80k_mapillary.pth) |
| SegFormer | InternImage-XL | 512x1024 | 330M | 2364G | [config](./configs/mapillary/segformer_internimage_xl_512x1024_80k_mapillary.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/segformer_internimage_xl_512x1024_80k_mapillary.pth) |
| Mask2Former | InternImage-H | 896x896 | 1094M | 7878G | [config](./configs/mapillary/mask2former_internimage_h_896x896_80k_mapillary.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/mask2former_internimage_h_896x896_80k_mapillary.pth) |
</div>
</details>
## Evaluation
To evaluate our `InternImage` on ADE20K val, run:
......
segmentation/configs/coco_stuff164k/README.md
View file @ b1238b9d
......@@ -10,6 +10,6 @@ The Common Objects in COntext-stuff (COCO-stuff) dataset is a dataset for scene
### Mask2Former + InternImage
| backbone | resolution | mIoU (ss) | train speed | train time | #param | FLOPs | Config | Download |
| :-----------: | :--------: | :-------: | :---------: | :--------: | :----: | :---: | :------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| InternImage-H | 896x896 | 52.6 | 1.6s / iter | 1.5d (2n) | 1.31B | 4635G | [config](./mask2former_internimage_h_896_80k_cocostuff164k.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/mask2former_internimage_h_896_80k_cocostuff164k.pth) \| [log](https://huggingface.co/OpenGVLab/InternImage/raw/main/mask2former_internimage_h_896_80k_cocostuff164k.log.json) |
| backbone | resolution | mIoU (ss/ms) | train speed | train time | #param | FLOPs | Config | Download |
| :-----------: | :--------: | :----------: | :---------: | :--------: | :----: | :---: | :------------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| InternImage-H | 896x896 | 52.6 / 52.8 | 1.6s / iter | 1.5d (2n) | 1.31B | 4635G | [config](./mask2former_internimage_h_896_80k_cocostuff164k.py) | [ckpt](https://huggingface.co/OpenGVLab/InternImage/resolve/main/mask2former_internimage_h_896_80k_cocostuff164k.pth) \| [log](https://huggingface.co/OpenGVLab/InternImage/raw/main/mask2former_internimage_h_896_80k_cocostuff164k.log.json) |
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment