Semi-Supervised Disparity Estimation with Deep Feature Reconstruction

06/01/2021

∙

by   Julia Guerrero-Viu, et al.

∙

0

∙

share

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.