Skip to content

Latest commit

 

History

History
111 lines (86 loc) · 31 KB

README.md

File metadata and controls

111 lines (86 loc) · 31 KB

FCN

Fully Convolutional Networks for Semantic Segmentation

Introduction

Official Repo

Code Snippet

Abstract

Convolutional networks are powerful visual models that yield hierarchies of features. We show that convolutional networks by themselves, trained end-to-end, pixels-to-pixels, exceed the state-of-the-art in semantic segmentation. Our key insight is to build "fully convolutional" networks that take input of arbitrary size and produce correspondingly-sized output with efficient inference and learning. We define and detail the space of fully convolutional networks, explain their application to spatially dense prediction tasks, and draw connections to prior models. We adapt contemporary classification networks (AlexNet, the VGG net, and GoogLeNet) into fully convolutional networks and transfer their learned representations by fine-tuning to the segmentation task. We then define a novel architecture that combines semantic information from a deep, coarse layer with appearance information from a shallow, fine layer to produce accurate and detailed segmentations. Our fully convolutional network achieves state-of-the-art segmentation of PASCAL VOC (20% relative improvement to 62.2% mean IU on 2012), NYUDv2, and SIFT Flow, while inference takes one third of a second for a typical image.

Results and models

Cityscapes

Method Backbone Crop Size Lr schd Mem (GB) Inf time (fps) Device mIoU mIoU(ms+flip) config download
FCN R-50-D8 512x1024 40000 5.7 4.17 V100 72.25 73.36 config model | log
FCN R-101-D8 512x1024 40000 9.2 2.66 V100 75.45 76.58 config model | log
FCN R-50-D8 769x769 40000 6.5 1.80 V100 71.47 72.54 config model | log
FCN R-101-D8 769x769 40000 10.4 1.19 V100 73.93 75.14 config model | log
FCN R-18-D8 512x1024 80000 1.7 14.65 V100 71.11 72.91 config model | log
FCN R-50-D8 512x1024 80000 - V100 73.61 74.24 config model | log
FCN R-101-D8 512x1024 80000 - - V100 75.13 75.94 config model | log
FCN (FP16) R-101-D8 512x1024 80000 5.37 8.64 V100 76.80 - config model | log
FCN R-18-D8 769x769 80000 1.9 6.40 V100 70.80 73.16 config model | log
FCN R-50-D8 769x769 80000 - - V100 72.64 73.32 config model | log
FCN R-101-D8 769x769 80000 - - V100 75.52 76.61 config model | log
FCN R-18b-D8 512x1024 80000 1.6 16.74 V100 70.24 72.77 config model | log
FCN R-50b-D8 512x1024 80000 5.6 4.20 V100 75.65 77.59 config model | log
FCN R-101b-D8 512x1024 80000 9.1 2.73 V100 77.37 78.77 config model | log
FCN R-18b-D8 769x769 80000 1.7 6.70 V100 69.66 72.07 config model | log
FCN R-50b-D8 769x769 80000 6.3 1.82 V100 73.83 76.60 config model | log
FCN R-101b-D8 769x769 80000 10.3 1.15 V100 77.02 78.67 config model | log
FCN (D6) R-50-D16 512x1024 40000 3.4 10.22 TITAN Xp 77.06 78.85 config model | log
FCN (D6) R-50-D16 512x1024 80000 - 10.35 TITAN Xp 77.27 78.88 config model | log
FCN (D6) R-50-D16 769x769 40000 3.7 4.17 TITAN Xp 76.82 78.22 config model | log
FCN (D6) R-50-D16 769x769 80000 - 4.15 TITAN Xp 77.04 78.40 config model | log
FCN (D6) R-101-D16 512x1024 40000 4.5 8.04 TITAN Xp 77.36 79.18 config model | log
FCN (D6) R-101-D16 512x1024 80000 - 8.26 TITAN Xp 78.46 80.42 config model | log
FCN (D6) R-101-D16 769x769 40000 5.0 3.12 TITAN Xp 77.28 78.95 config model | log
FCN (D6) R-101-D16 769x769 80000 - 3.21 TITAN Xp 78.06 79.58 config model | log
FCN (D6) R-50b-D16 512x1024 80000 3.2 10.16 TITAN Xp 76.99 79.03 config model | log
FCN (D6) R-50b-D16 769x769 80000 3.6 4.17 TITAN Xp 76.86 78.52 config model | log
FCN (D6) R-101b-D16 512x1024 80000 4.3 8.46 TITAN Xp 77.72 79.53 config model | log
FCN (D6) R-101b-D16 769x769 80000 4.8 3.32 TITAN Xp 77.34 78.91 config model | log

ADE20K

Method Backbone Crop Size Lr schd Mem (GB) Inf time (fps) Device mIoU mIoU(ms+flip) config download
FCN R-50-D8 512x512 80000 8.5 23.49 V100 35.94 37.94 config model | log
FCN R-101-D8 512x512 80000 12 14.78 V100 39.61 40.83 config model | log
FCN R-50-D8 512x512 160000 - - V100 36.10 38.08 config model | log
FCN R-101-D8 512x512 160000 - - V100 39.91 41.40 config model | log

Pascal VOC 2012 + Aug

Method Backbone Crop Size Lr schd Mem (GB) Inf time (fps) Device mIoU mIoU(ms+flip) config download
FCN R-50-D8 512x512 20000 5.7 23.28 V100 67.08 69.94 config model | log
FCN R-101-D8 512x512 20000 9.2 14.81 V100 71.16 73.57 config model | log
FCN R-50-D8 512x512 40000 - - V100 66.97 69.04 config model | log
FCN R-101-D8 512x512 40000 - - V100 69.91 72.38 config model | log

Pascal Context

Method Backbone Crop Size Lr schd Mem (GB) Inf time (fps) Device mIoU mIoU(ms+flip) config download
FCN R-101-D8 480x480 40000 - 9.93 V100 44.43 45.63 config model | log
FCN R-101-D8 480x480 80000 - - V100 44.13 45.26 config model | log

Pascal Context 59

Method Backbone Crop Size Lr schd Mem (GB) Inf time (fps) Device mIoU mIoU(ms+flip) config download
FCN R-101-D8 480x480 40000 - - V100 48.42 50.4 config model | log
FCN R-101-D8 480x480 80000 - - V100 49.35 51.38 config model | log

Note:

  • FP16 means Mixed Precision (FP16) is adopted in training.
  • FCN D6 means dilation rate of convolution operator in FCN is 6.

Citation

@article{shelhamer2017fully,
  title={Fully convolutional networks for semantic segmentation},
  author={Shelhamer, Evan and Long, Jonathan and Darrell, Trevor},
  journal={IEEE transactions on pattern analysis and machine intelligence},
  volume={39},
  number={4},
  pages={640--651},
  year={2017},
  publisher={IEEE Trans Pattern Anal Mach Intell}
}