Trajectory Densification and Depth from Perspective-based Blur
This work addresses depth estimation and trajectory densification for computer vision applications, particularly in handheld or unstabilized camera scenarios, representing an incremental advancement.
The paper tackles the problem of estimating metric depth and densifying sparse trajectories from videos affected by perspective-based blur due to camera rotation, achieving strong performance with superior precision in handheld shooting settings.
In the absence of a mechanical stabilizer, the camera undergoes inevitable rotational dynamics during capturing, which induces perspective-based blur especially under long-exposure scenarios. From an optical standpoint, perspective-based blur is depth-position-dependent: objects residing at distinct spatial locations incur different blur levels even under the same imaging settings. Inspired by this, we propose a novel method that estimate metric depth by examining the blur pattern of a video stream and dense trajectory via joint optical design algorithm. Specifically, we employ off-the-shelf vision encoder and point tracker to extract video information. Then, we estimate depth map via windowed embedding and multi-window aggregation, and densify the sparse trajectory from the optical algorithm using a vision-language model. Evaluations on multiple depth datasets demonstrate that our method attains strong performance over large depth range, while maintaining favorable generalization. Relative to the real trajectory in handheld shooting settings, our optical algorithm achieves superior precision and the dense reconstruction maintains strong accuracy.