We propose a simple yet efficient anchor-free instance segmentation, called CenterMask, that adds a novel spatial attention-guided mask (SAG-Mask) branch to anchor-free one stage object detector (FCOS) in the same vein with Mask R-CNN. Plugged into the FCOS object detector, the SAG-Mask branch predicts a segmentation mask on each box with the spatial attention map that helps to focus on informative pixels and suppress noise. We also present an improved VoVNetV2 with two effective strategies: adds (1) residual connection for alleviating the saturation problem of larger VoVNet and (2) effective Squeeze-Excitation (eSE) deals with the information loss problem of original SE. With SAG-Mask and VoVNetV2, we deign CenterMask and CenterMask-Lite that are targeted to large and small models, respectively. CenterMask outperforms all previous state-of-the-art models at a much faster speed. CenterMask-Lite also achieves 33.4% mask AP / 38.0% box AP, outperforming YOLACT by 2.6 / 7.0 AP gain, respectively, at over 35fps on Titan Xp. We hope that CenterMask and VoVNetV2 can serve as a solid baseline of real-time instance segmentation and backbone network for various vision tasks, respectively.

• Open the official repo and code will be released after refactoring. (05/12/2019)
• Release code and MobileNetV2 & ResNet backbone models shown in the [paper].(10/12/2019)

• The release of VoVNetV2 models is delayed due to the internal affairs concerning patent.

• V100 or Titan Xp GPU
• CUDA 10.0
• cuDNN7.3
• pytorch1.1
• Implemented on fcos and maskrcn-benchmark

Note that RetinaMask, Mask R-CNN, and CenterMask are implemented by using same baseline code(maskrcnn-benchmark) and all models are trained using multi-scale training augmentation.

We expect that if we implement our CenterMask based on detectron2, it will get better performance.

Check INSTALL.md for installation instructions which is orginate from maskrcnn-benchmark.

Follow the instructions of maskrcnn-benchmark guides.

If you want multi-gpu (e.g.,8) training,

export NGPUS=8
python -m torch.distributed.launch --nproc_per_node=$NGPUS tools/train_net.py --config-file "configs/centermask/centermask_R_50_FPN_1x.yaml" 

Follow the instruction of maskrcnn-benchmark

First of all, you have to download the weight file you want to inference.

For examaple (CenterMask-Lite-R-50),

wget https://dl.dropbox.com/s/2enqxenccz4xy6l/centermask-lite-R-50-ms-bs32-1x.pth
export NGPUS=8
python -m torch.distributed.launch --nproc_per_node=$NGPUS tools/test_net.py --config-file "configs/centermask/centermask_R_50_FPN_lite_res600_ms_bs32_1x.yaml"   TEST.IMS_PER_BATCH 16 MODEL.WEIGHT centermask-lite-R-50-ms-bs32-1x.pth

wget https://dl.dropbox.com/s/2enqxenccz4xy6l/centermask-lite-R-50-ms-bs32-1x.pth
CUDA_VISIBLE_DEVICES=0
python tools/test_net.py --config-file "configs/centermask/centermask_R_50_FPN_lite_res600_ms_bs32_1x.yaml" TEST.IMS_PER_BATCH 1 MODEL.WEIGHT centermask-lite-R-50-ms-bs32-1x.pth

• train-time augmentation + 3x schedule for comparing with detectron2 models
• quick-demo
• ResNet-50 & ResNeXt-101-32x8d
• arxiv paper update
• VoVNetV2 backbones

Please cite our paper in your publications if it helps your research:

@article{lee2019centermask,
  title={CenterMask: Real-Time Anchor-Free Instance Segmentation},
  author={Lee, Youngwan and Park, Jongyoul},
  journal={arXiv preprint arXiv:1911.06667},
  year={2019}
}