SelFlow: Self-Supervised Learning of Optical Flow

CVPR 2019 Pengpeng LiuMichael LyuIrwin KingJia Xu

We present a self-supervised learning approach for optical flow. Our method distills reliable flow estimations from non-occluded pixels, and uses these predictions as ground truth to learn optical flow for hallucinated occlusions... (read more)

PDF Abstract

Evaluation results from the paper

Task Dataset Model Metric name Metric value Global rank Compare
Optical Flow Estimation KITTI 2012 SelFlow Average End-Point Error 1.5 # 1
Optical Flow Estimation KITTI 2015 SelFlow Fl-all 8.42 # 3
Optical Flow Estimation Sintel-clean SelFlow Average End-Point Error 3.74 # 3
Optical Flow Estimation Sintel-final SelFlow Average End-Point Error 4.26 # 1