deeplabv3plus

Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation

Introduction

@inproceedings{deeplabv3plus2018,
  title={Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation},
  author={Liang-Chieh Chen and Yukun Zhu and George Papandreou and Florian Schroff and Hartwig Adam},
  booktitle={ECCV},
  year={2018}
}

Results and models

Note: D-8 here corresponding to the output stride 8 setting for DeepLab series.

Cityscapes

Method	Backbone	Crop Size	Lr schd	Mem (GB)	Inf time (fps)	mIoU	mIoU(ms+flip)	download
DeepLabV3+	R-50-D8	512x1024	40000	7.5	3.94	79.61	81.01	model | log
DeepLabV3+	R-101-D8	512x1024	40000	11	2.60	80.21	81.82	model | log
DeepLabV3+	R-50-D8	769x769	40000	8.5	1.72	78.97	80.46	model | log
DeepLabV3+	R-101-D8	769x769	40000	12.5	1.15	79.46	80.50	model | log
DeepLabV3+	R-50-D8	512x1024	80000	-	-	80.09	81.13	model | log
DeepLabV3+	R-101-D8	512x1024	80000	-	-	80.97	82.03	model | log
DeepLabV3+	R-50-D8	769x769	80000	-	-	79.83	81.48	model | log
DeepLabV3+	R-101-D8	769x769	80000	-	-	80.98	82.18	model | log

ADE20K

Method	Backbone	Crop Size	Lr schd	Mem (GB)	Inf time (fps)	mIoU	mIoU(ms+flip)	download
DeepLabV3+	R-50-D8	512x512	80000	10.6	21.01	42.72	43.75	model | log
DeepLabV3+	R-101-D8	512x512	80000	14.1	14.16	44.60	46.06	model | log
DeepLabV3+	R-50-D8	512x512	160000	-	-	43.95	44.93	model | log
DeepLabV3+	R-101-D8	512x512	160000	-	-	45.47	46.35	model | log

Pascal VOC 2012 + Aug

Method	Backbone	Crop Size	Lr schd	Mem (GB)	Inf time (fps)	mIoU	mIoU(ms+flip)	download
DeepLabV3+	R-50-D8	512x512	20000	7.6	21	75.93	77.50	model | log
DeepLabV3+	R-101-D8	512x512	20000	11	13.88	77.22	78.59	model | log
DeepLabV3+	R-50-D8	512x512	40000	-	-	76.81	77.57	model | log
DeepLabV3+	R-101-D8	512x512	40000	-	-	78.62	79.53	model | log