Automatic generation of dense non-rigid optical flow

作者：

Highlights：

• First method to automatically generate dense non-rigid optical flow data for training.

• A large optical flow dataset with natural textures and non-rigid motion created from DAVIS videos.

• Evaluating optical flow networks trained using flow references from various methods.

摘要

•First method to automatically generate dense non-rigid optical flow data for training.•A large optical flow dataset with natural textures and non-rigid motion created from DAVIS videos.•Evaluating optical flow networks trained using flow references from various methods.

论文关键词：non-rigi,optical flow,dataset,generation,as-rigid-as-possible

论文评审过程：Received 1 August 2020, Revised 31 August 2021, Accepted 4 September 2021, Available online 8 September 2021, Version of Record 15 September 2021.

论文官网地址：https://doi.org/10.1016/j.cviu.2021.103274