Sparse-RCNN inplementation

This is an unofficial pytorch implementation of SparseRCNN object detection as described in Sparse R-CNN: End-to-End Object Detection with Learnable Proposals by Peize Sun, Rufeng Zhang, Yi Jiang, Tao Kong, Chenfeng Xu, Wei Zhan, Masayoshi Tomizuka, Lei Li, Zehuan Yuan, Changhu Wang, Ping Luo

Result

I only train this model with resnet50 backbone on the coco dataset for 6 epochs, and the default max epochs is 30.

AP : 35.747, AP50 : 53.073, AP75 : 38.319

Roadmap

• Use albumentations instead of the basic transforms
• Add eval script and demo
• fp16 mixed precision training
• MAE
• Voc dataset support
• Support for multiple GPUs

Example

• train on coco dataset with resnet50 backbone

  python train.py --set BASE_ROOT /home/input/coco-2017-dataset/coco2017 SOLVER.IMS_PER_BATCH 4 MODEL.BACKBONE "resnet50"
  

• train on coco dataset using fp16 mixed precision training with efficientnet_b3 backbone

  python train.py \
    --fp16_mix \
    --set BASE_ROOT /home/input/coco-2017-dataset/coco2017 SOLVER.IMS_PER_BATCH 4 MODEL.BACKBONE "efficientnet_b3"
  

• or you can specify the backbone and dataset in config file

  python train.py \
    --fp16_mix \
    --dataset sparse_rcnn/configs/coco.yaml \
    --model configs/sparse_rcnn.yaml
  

Reference

original official implement based on detectron2 and DETR

@article{peize2020sparse,
  title   =  {{SparseR-CNN}: End-to-End Object Detection with Learnable Proposals},
  author  =  {Peize Sun and Rufeng Zhang and Yi Jiang and Tao Kong and Chenfeng Xu and Wei Zhan and Masayoshi Tomizuka and Lei Li and Zehuan Yuan and Changhu Wang and Ping Luo},
  journal =  {arXiv preprint arXiv:2011.12450},
  year    =  {2020}
}