English

Semi-Supervised Disparity Estimation with Deep Feature Reconstruction

Computer Vision and Pattern Recognition 2021-06-02 v1

Abstract

Despite the success of deep learning in disparity estimation, the domain generalization gap remains an issue. We propose a semi-supervised pipeline that successfully adapts DispNet to a real-world domain by joint supervised training on labeled synthetic data and self-supervised training on unlabeled real data. Furthermore, accounting for the limitations of the widely-used photometric loss, we analyze the impact of deep feature reconstruction as a promising supervisory signal for disparity estimation.

Keywords

Cite

@article{arxiv.2106.00318,
  title  = {Semi-Supervised Disparity Estimation with Deep Feature Reconstruction},
  author = {Julia Guerrero-Viu and Sergio Izquierdo and Philipp Schröppel and Thomas Brox},
  journal= {arXiv preprint arXiv:2106.00318},
  year   = {2021}
}

Comments

Women in Computer Vision workshop CVPR 2021

R2 v1 2026-06-24T02:41:53.064Z