You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

README.md 8.2 kB

2 years ago
12345678910111213141516171819202122232425262728293031323334353637383940
  1. # Focal Loss for Dense Object Detection
  2. ## Introduction
  3. <!-- [ALGORITHM] -->
  4. ```latex
  5. @inproceedings{lin2017focal,
  6. title={Focal loss for dense object detection},
  7. author={Lin, Tsung-Yi and Goyal, Priya and Girshick, Ross and He, Kaiming and Doll{\'a}r, Piotr},
  8. booktitle={Proceedings of the IEEE international conference on computer vision},
  9. year={2017}
  10. }
  11. ```
  12. ## Results and models
  13. | Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | Config | Download |
  14. | :-------------: | :-----: | :-----: | :------: | :------------: | :----: | :------: | :--------: |
  15. | R-50-FPN | caffe | 1x | 3.5 | 18.6 | 36.3 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r50_caffe_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_caffe_fpn_1x_coco/retinanet_r50_caffe_fpn_1x_coco_20200531-f11027c5.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_caffe_fpn_1x_coco/retinanet_r50_caffe_fpn_1x_coco_20200531_012518.log.json) |
  16. | R-50-FPN | pytorch | 1x | 3.8 | 19.0 | 36.5 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r50_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_fpn_1x_coco/retinanet_r50_fpn_1x_coco_20200130-c2398f9e.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_fpn_1x_coco/retinanet_r50_fpn_1x_coco_20200130_002941.log.json) |
  17. | R-50-FPN | pytorch | 2x | - | - | 37.4 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r50_fpn_2x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_fpn_2x_coco/retinanet_r50_fpn_2x_coco_20200131-fdb43119.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_fpn_2x_coco/retinanet_r50_fpn_2x_coco_20200131_114738.log.json) |
  18. | R-101-FPN | caffe | 1x | 5.5 | 14.7 | 38.5 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r101_caffe_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_caffe_fpn_1x_coco/retinanet_r101_caffe_fpn_1x_coco_20200531-b428fa0f.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_caffe_fpn_1x_coco/retinanet_r101_caffe_fpn_1x_coco_20200531_012536.log.json) |
  19. | R-101-FPN | pytorch | 1x | 5.7 | 15.0 | 38.5 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r101_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_fpn_1x_coco/retinanet_r101_fpn_1x_coco_20200130-7a93545f.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_fpn_1x_coco/retinanet_r101_fpn_1x_coco_20200130_003055.log.json) |
  20. | R-101-FPN | pytorch | 2x | - | - | 38.9 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r101_fpn_2x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_fpn_2x_coco/retinanet_r101_fpn_2x_coco_20200131-5560aee8.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_fpn_2x_coco/retinanet_r101_fpn_2x_coco_20200131_114859.log.json) |
  21. | X-101-32x4d-FPN | pytorch | 1x | 7.0 | 12.1 | 39.9 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_x101_32x4d_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_32x4d_fpn_1x_coco/retinanet_x101_32x4d_fpn_1x_coco_20200130-5c8b7ec4.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_32x4d_fpn_1x_coco/retinanet_x101_32x4d_fpn_1x_coco_20200130_003004.log.json) |
  22. | X-101-32x4d-FPN | pytorch | 2x | - | - | 40.1 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_x101_32x4d_fpn_2x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_32x4d_fpn_2x_coco/retinanet_x101_32x4d_fpn_2x_coco_20200131-237fc5e1.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_32x4d_fpn_2x_coco/retinanet_x101_32x4d_fpn_2x_coco_20200131_114812.log.json) |
  23. | X-101-64x4d-FPN | pytorch | 1x | 10.0 | 8.7 | 41.0 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_x101_64x4d_fpn_1x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_64x4d_fpn_1x_coco/retinanet_x101_64x4d_fpn_1x_coco_20200130-366f5af1.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_64x4d_fpn_1x_coco/retinanet_x101_64x4d_fpn_1x_coco_20200130_003008.log.json) |
  24. | X-101-64x4d-FPN | pytorch | 2x | - | - | 40.8 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_x101_64x4d_fpn_2x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_64x4d_fpn_2x_coco/retinanet_x101_64x4d_fpn_2x_coco_20200131-bca068ab.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_64x4d_fpn_2x_coco/retinanet_x101_64x4d_fpn_2x_coco_20200131_114833.log.json) |
  25. ## Pre-trained Models
  26. We also train some models with longer schedules and multi-scale training. The users could finetune them for downstream tasks.
  27. | Backbone | Style | Lr schd | Mem (GB) | box AP | Config | Download |
  28. | :----------------: | :-----: | :-----: | :------: | :----: | :------: | :--------: |
  29. | R-50-FPN | pytorch| 3x | 3.5 | 39.5 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r50_fpn_mstrain_640-800_3x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_fpn_mstrain_3x_coco/retinanet_r50_fpn_mstrain_3x_coco_20210718_220633-88476508.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r50_fpn_mstrain_3x_coco/retinanet_r50_fpn_mstrain_3x_coco_20210718_220633-88476508.log.json)
  30. | R-101-FPN | caffe | 3x | 5.4 | 40.7 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r101_caffe_fpn_mstrain_3x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_caffe_fpn_mstrain_3x_coco/retinanet_r101_caffe_fpn_mstrain_3x_coco_20210721_063439-88a8a944.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_caffe_fpn_mstrain_3x_coco/retinanet_r101_caffe_fpn_mstrain_3x_coco_20210721_063439-88a8a944.log.json)
  31. | R-101-FPN | pytorch| 3x | 5.4 | 41 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_r101_fpn_mstrain_640-800_3x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_fpn_mstrain_3x_coco/retinanet_r101_fpn_mstrain_3x_coco_20210720_214650-7ee888e0.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_r101_fpn_mstrain_3x_coco/retinanet_r101_fpn_mstrain_3x_coco_20210720_214650-7ee888e0.log.json)
  32. | X-101-64x4d-FPN | pytorch| 3x | 9.8 | 41.6 | [config](https://github.com/open-mmlab/mmdetection/tree/master/configs/retinanet/retinanet_x101_64x4d_fpn_mstrain_640-800_3x_coco.py) | [model](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_64x4d_fpn_mstrain_3x_coco/retinanet_x101_64x4d_fpn_mstrain_3x_coco_20210719_051838-022c2187.pth) &#124; [log](https://download.openmmlab.com/mmdetection/v2.0/retinanet/retinanet_x101_64x4d_fpn_mstrain_3x_coco/retinanet_x101_64x4d_fpn_mstrain_3x_coco_20210719_051838-022c2187.log.json)

No Description

Contributors (3)