Edge-aware Consistent Stereo Video Depth Estimation
This work addresses depth estimation for applications like scene reconstruction and augmented reality, but it is incremental as it builds on existing monocular and stereo techniques.
The paper tackles the problem of dense video depth estimation by proposing a method for stereo videos that uses left-right consistency loss and an edge-preserving loss to reduce flickering and improve detail visibility, achieving accurate depth maps.
Video depth estimation is crucial in various applications, such as scene reconstruction and augmented reality. In contrast to the naive method of estimating depths from images, a more sophisticated approach uses temporal information, thereby eliminating flickering and geometrical inconsistencies. We propose a consistent method for dense video depth estimation; however, unlike the existing monocular methods, ours relates to stereo videos. This technique overcomes the limitations arising from the monocular input. As a benefit of using stereo inputs, a left-right consistency loss is introduced to improve the performance. Besides, we use SLAM-based camera pose estimation in the process. To address the problem of depth blurriness during test-time training (TTT), we present an edge-preserving loss function that improves the visibility of fine details while preserving geometrical consistency. We show that our edge-aware stereo video model can accurately estimate the dense depth maps.