Monocular contextual constraint for stereo matching with adaptive weights assignment

作者：

Highlights：

• A co-learning framework named CLStereo with monocular and stereo branches is proposed

• We introduce contextual constraints to transfer monocular prior knowledge

• We propose an adaptive weights assignment to balance the co-learning of both branches

• Comparisons with state-of-the-art methods demonstrate the superiority of our method

摘要

•A co-learning framework named CLStereo with monocular and stereo branches is proposed•We introduce contextual constraints to transfer monocular prior knowledge•We propose an adaptive weights assignment to balance the co-learning of both branches•Comparisons with state-of-the-art methods demonstrate the superiority of our method

论文关键词：Deep learning,Stereo matching,Monocular contextual constraint,Adaptive weights assignment

论文评审过程：Received 14 December 2021, Revised 22 February 2022, Accepted 23 February 2022, Available online 26 February 2022, Version of Record 9 March 2022.

论文官网地址：https://doi.org/10.1016/j.imavis.2022.104424