Semi-Supervised Disparity Estimation with Deep Feature Reconstruction
This work addresses domain generalization for disparity estimation, which is important for applications like robotics and autonomous driving, but it appears incremental as it builds on existing methods like DispNet and photometric loss.
The paper tackles the domain generalization gap in deep learning-based disparity estimation by proposing a semi-supervised pipeline that adapts DispNet to real-world domains using joint supervised training on synthetic data and self-supervised training on unlabeled real data, and analyzes deep feature reconstruction as a supervisory signal to address limitations of photometric loss.
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.