Add unified PEMOLA model weight overview
Browse files
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- pemola
|
| 4 |
+
- panoptic-segmentation
|
| 5 |
+
- occlusion
|
| 6 |
+
- icme-2026
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# PEMOLA (ICME 2026)
|
| 10 |
+
|
| 11 |
+
Weights and matching configurations for the **PEMOLA** paper at **ICME 2026**. The results below are reported in the paper. Select a model name for its setup and usage instructions.
|
| 12 |
+
|
| 13 |
+
| Model | Dataset | Result | Config / Weights |
|
| 14 |
+
| :--- | :--- | :---: | :---: |
|
| 15 |
+
| [Mask2Former R50](https://huggingface.co/weiwb/Mask2Former-R50-COCO-OLAC) | COCO-OLAC | PQ 40.7 | [config](https://huggingface.co/weiwb/Mask2Former-R50-COCO-OLAC/resolve/main/config.yaml) / [weights](https://huggingface.co/weiwb/Mask2Former-R50-COCO-OLAC/resolve/main/model_final.pth) |
|
| 16 |
+
| [PEMOLA + Mask2Former R50](https://huggingface.co/weiwb/PEMOLA-Mask2Former-R50-COCO-OLAC) | COCO-OLAC | PQ 41.5 | [config](https://huggingface.co/weiwb/PEMOLA-Mask2Former-R50-COCO-OLAC/resolve/main/config.yaml) / [weights](https://huggingface.co/weiwb/PEMOLA-Mask2Former-R50-COCO-OLAC/resolve/main/model_final.pth) |
|
| 17 |
+
| [Mask2Former R50](https://huggingface.co/weiwb/Mask2Former-R50-Cityscapes-OLAC) | Cityscapes-OLAC | PQ 61.5 | [config](https://huggingface.co/weiwb/Mask2Former-R50-Cityscapes-OLAC/resolve/main/config.yaml) / [weights](https://huggingface.co/weiwb/Mask2Former-R50-Cityscapes-OLAC/resolve/main/model_final.pth) |
|
| 18 |
+
| [PEMOLA + Mask2Former R50](https://huggingface.co/weiwb/PEMOLA-Mask2Former-R50-Cityscapes-OLAC) | Cityscapes-OLAC | PQ 62.3 | [config](https://huggingface.co/weiwb/PEMOLA-Mask2Former-R50-Cityscapes-OLAC/resolve/main/config.yaml) / [weights](https://huggingface.co/weiwb/PEMOLA-Mask2Former-R50-Cityscapes-OLAC/resolve/main/model_final.pth) |
|
| 19 |
+
| [Mask DINO R50](https://huggingface.co/weiwb/MaskDINO-R50-COCO-OLAC) | COCO-OLAC | PQ 44.0 | [config](https://huggingface.co/weiwb/MaskDINO-R50-COCO-OLAC/resolve/main/config.yaml) / [weights](https://huggingface.co/weiwb/MaskDINO-R50-COCO-OLAC/resolve/main/model_final.pth) |
|
| 20 |
+
| [PEMOLA + Mask DINO R50](https://huggingface.co/weiwb/PEMOLA-MaskDINO-R50-COCO-OLAC) | COCO-OLAC | PQ 44.8 | [config](https://huggingface.co/weiwb/PEMOLA-MaskDINO-R50-COCO-OLAC/resolve/main/config.yaml) / [weights](https://huggingface.co/weiwb/PEMOLA-MaskDINO-R50-COCO-OLAC/resolve/main/model_final.pth) |
|
| 21 |
+
| [Occlusion classifier: Swin-L 384](https://huggingface.co/weiwb/PEMOLA-Occlusion-Swin-L-384) | COCO-OLAC test, black background | Top-1 75.3% | [config](https://huggingface.co/weiwb/PEMOLA-Occlusion-Swin-L-384/resolve/main/config.yaml) / [weights](https://huggingface.co/weiwb/PEMOLA-Occlusion-Swin-L-384/resolve/main/ep29.pth) |
|
| 22 |
+
|
| 23 |
+
For source code, [installation](https://github.com/wenbo-wei/PEMOLA#installation), and [inference](https://github.com/wenbo-wei/PEMOLA#inference--visualization), see the [PEMOLA GitHub repository](https://github.com/wenbo-wei/PEMOLA). Mask DINO models use the dedicated [Mask DINO implementation](https://github.com/wenbo-wei/MaskDINO).
|
| 24 |
+
|
| 25 |
+
Use the configuration shipped with each checkpoint. Each linked model repository provides its own license and usage details.
|