README.md

GCNet for Object Detection

By Yue Cao, Jiarui Xu, Stephen Lin, Fangyun Wei, Han Hu.

We provide config files to reproduce the results in the paper for "GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond" on COCO object detection.

Introduction

GCNet is initially described in arxiv. Via absorbing advantages of Non-Local Networks (NLNet) and Squeeze-Excitation Networks (SENet), GCNet provides a simple, fast and effective approach for global context modeling, which generally outperforms both NLNet and SENet on major benchmarks for various recognition tasks.

Citing GCNet

@article{cao2019GCNet,
  title={GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond},
  author={Cao, Yue and Xu, Jiarui and Lin, Stephen and Wei, Fangyun and Hu, Han},
  journal={arXiv preprint arXiv:1904.11492},
  year={2019}
}

Results and models

The results on COCO 2017val are shown in the below table.

Backbone	Model	Context	Lr schd	Mem (GB)	Inf time (fps)	box AP	mask AP	Download
R-50-FPN	Mask	GC(c3-c5, r16)	1x	5.0		39.7	35.9	model | log
R-50-FPN	Mask	GC(c3-c5, r4)	1x	5.1	15.0	39.9	36.0	model | log
R-101-FPN	Mask	GC(c3-c5, r16)	1x	7.6	11.4	41.3	37.2	model | log
R-101-FPN	Mask	GC(c3-c5, r4)	1x	7.8	11.6	42.2	37.8	model | log

Backbone	Model	Context	Lr schd	Mem (GB)	Inf time (fps)	box AP	mask AP	Download
R-50-FPN	Mask	-	1x	4.4	16.6	38.4	34.6	model | log
R-50-FPN	Mask	GC(c3-c5, r16)	1x	5.0	15.5	40.4	36.2	model | log
R-50-FPN	Mask	GC(c3-c5, r4)	1x	5.1	15.1	40.7	36.5	model | log
R-101-FPN	Mask	-	1x	6.4	13.3	40.5	36.3	model | log
R-101-FPN	Mask	GC(c3-c5, r16)	1x	7.6	12.0	42.2	37.8	model | log
R-101-FPN	Mask	GC(c3-c5, r4)	1x	7.8	11.8	42.2	37.8	model | log
X-101-FPN	Mask	-	1x	7.6	11.3	42.4	37.7	model | log
X-101-FPN	Mask	GC(c3-c5, r16)	1x	8.8	9.8	43.5	38.6	model | log
X-101-FPN	Mask	GC(c3-c5, r4)	1x	9.0	9.7	43.9	39.0	model | log
X-101-FPN	Cascade Mask	-	1x	9.2	8.4	44.7	38.6	model | log
X-101-FPN	Cascade Mask	GC(c3-c5, r16)	1x	10.3	7.7	46.2	39.7	modle | log
X-101-FPN	Cascade Mask	GC(c3-c5, r4)	1x			45.9	39.6
X-101-FPN	DCN Cascade Mask	-	1x			44.9	38.9	modle | log
X-101-FPN	DCN Cascade Mask	GC(c3-c5, r16)	1x			44.6		modle | log
X-101-FPN	DCN Cascade Mask	GC(c3-c5, r4)	1x			45.7	39.5	modle | log

Notes:

• The SyncBN is added in the backbone for all models in Table 2.
• GC denotes Global Context (GC) block is inserted after 1x1 conv of backbone.
• DCN denotes replace 3x3 conv with 3x3 Deformable Convolution in c3-c5 stages of backbone.
• r4 and r16 denote ratio 4 and ratio 16 in GC block respectively.