README.md

Deformable Convolutional Networks

Introduction

@inproceedings{dai2017deformable,
  title={Deformable Convolutional Networks},
  author={Dai, Jifeng and Qi, Haozhi and Xiong, Yuwen and Li, Yi and Zhang, Guodong and Hu, Han and Wei, Yichen},
  booktitle={Proceedings of the IEEE international conference on computer vision},
  year={2017}
}

@article{zhu2018deformable,
  title={Deformable ConvNets v2: More Deformable, Better Results},
  author={Zhu, Xizhou and Hu, Han and Lin, Stephen and Dai, Jifeng},
  journal={arXiv preprint arXiv:1811.11168},
  year={2018}
}

Results and Models

Backbone	Model	Style	Conv	Pool	Lr schd	Mem (GB)	Train time (s/iter)	Inf time (fps)	box AP	mask AP	Download
R-50-FPN	Faster	pytorch	dconv(c3-c5)	-	1x	3.9	0.594	10.2	40.0	-	model
R-50-FPN	Faster	pytorch	mdconv(c3-c5)	-	1x	3.7	0.598	10.0	40.2	-	model
*R-50-FPN (dg=4)	Faster	pytorch	mdconv(c3-c5)	-	1x	3.9	0.350	11.36	40.4	-	model
R-50-FPN	Faster	pytorch	-	dpool	1x	4.6	0.714	8.7	37.8	-	model
R-50-FPN	Faster	pytorch	-	mdpool	1x	5.2	0.769	8.2	38.0	-	model
R-101-FPN	Faster	pytorch	dconv(c3-c5)	-	1x	5.8	0.811	8.0	42.1	-	model
X-101-32x4d-FPN	Faster	pytorch	dconv(c3-c5)	-	1x	7.1	1.126	6.6	43.4	-	model
R-50-FPN	Mask	pytorch	dconv(c3-c5)	-	1x	4.5	0.712	7.7	41.1	37.2	model
R-50-FPN	Mask	pytorch	mdconv(c3-c5)	-	1x	4.5	0.712	7.7	41.3	37.3	model
R-101-FPN	Mask	pytorch	dconv(c3-c5)	-	1x	6.4	0.939	6.5	43.2	38.7	model
R-50-FPN	Cascade	pytorch	dconv(c3-c5)	-	1x	4.4	0.660	7.6	44.0	-	model
R-101-FPN	Cascade	pytorch	dconv(c3-c5)	-	1x	6.3	0.881	6.8	45.0	-	model
R-50-FPN	Cascade Mask	pytorch	dconv(c3-c5)	-	1x	6.6	0.942	5.7	44.4	38.3	model
R-101-FPN	Cascade Mask	pytorch	dconv(c3-c5)	-	1x	8.5	1.156	5.1	45.7	39.4	model

Notes:

• dconv and mdconv denote (modulated) deformable convolution, c3-c5 means adding dconv in resnet stage 3 to 5. dpool and mdpool denote (modulated) deformable roi pooling.
• The dcn ops are modified from https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch, which should be more memory efficient and slightly faster.
• (*) For R-50-FPN (dg=4), dg is short for deformable_group. This model is trained and tested on Amazon EC2 p3dn.24xlarge instance.
• Memory, Train/Inf time is outdated.