FDFlowNet: Fast Optical Flow Estimation using a Deep Lightweight Network

22 Jun 2020  ·  Lingtong Kong, Jie Yang ·

Significant progress has been made for estimating optical flow using deep neural networks. Advanced deep models achieve accurate flow estimation often with a considerable computation complexity and time-consuming training processes. In this work, we present a lightweight yet effective model for real-time optical flow estimation, termed FDFlowNet (fast deep flownet). We achieve better or similar accuracy on the challenging KITTI and Sintel benchmarks while being about 2 times faster than PWC-Net. This is achieved by a carefully-designed structure and newly proposed components. We first introduce an U-shape network for constructing multi-scale feature which benefits upper levels with global receptive field compared with pyramid network. In each scale, a partial fully connected structure with dilated convolution is proposed for flow estimation that obtains a good balance among speed, accuracy and number of parameters compared with sequential connected and dense connected structures. Experiments demonstrate that our model achieves state-of-the-art performance while being fast and lightweight.

PDF Abstract
No code implementations yet. Submit your code now
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Optical Flow Estimation KITTI 2012 FDFlowNet-ft Average End-Point Error 1.5 # 7
Optical Flow Estimation Sintel-clean FDFlowNet-ft Average End-Point Error 3.71 # 20
Optical Flow Estimation Sintel-final FDFlowNet-ft Average End-Point Error 5.11 # 22

Methods