Add MaskRCNN improved weights

[datumbox]

Fixes #5307

Adds new pre-trained weights for MaskRCNN + ResNet50 + FPN for the v2 variant with post-paper optimizations (no FrozenBN + c5 instead of p5 input on extra layers + heavier RPN/Box/Mask Heads with BNs). It improves the previous baseline by +9.5 mAP for boxes and +7.2 mAP for masks.

Trained with:

python -u run_with_submitit.py --ngpus 8 --nodes 4 --dataset coco --model maskrcnn_resnet50_fpn_v2 \
--epochs 400 --lr-steps 352 384 --lr 0.1 --batch-size 2 --weight-decay 0.00004 --sync-bn \
--data-augmentation lsj

Verified with:

torchrun --nproc_per_node=1 train.py --test-only --weights MaskRCNN_ResNet50_FPN_V2_Weights.COCO_V1 \
--model maskrcnn_resnet50_fpn_v2 -b 1

IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.474
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.679
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.518
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.308
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.512
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.609
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.367
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.589
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.618
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.443
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.652
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.762
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.418
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.651
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.450
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.222
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.447
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.602
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.337
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.527
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.551
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.363
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.587
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.715

[YosuaMichael]

LGTM!

[dscho15]

Hey, where can i find the augmentations used for training this model?

[NicolasHug]

@dscho15 The model was trained with the command above in #5773 (comment). It refers to these scripts, where you'll find the corresponding code https://github.com/pytorch/vision/tree/main/references/detection