×
Apr 18, 2023 · In this paper, we propose a novel method to learn to fuse the multi-view and monocular cues encoded as volumes without needing the heuristically crafted masks.
Focusing on the dynamic scenes, we first analyze the be- haviors of the depth estimation methods relying on multi- view and monocular cues, as in Sec. 3.1.
In this paper, we propose a novel method to learn to fuse the multi-view and monocular cues encoded as volumes without needing the heuristically crafted masks.
We aim to propagate the multi-frame static (yellow box) depth to the monocular cues and let monocular cues in dynamic areas (red box) enhance the multi-frame ...
Focusing on the dynamic scenes, we first analyze the be- haviors of the depth estimation methods relying on multi- view and monocular cues, as in Sec. 3.1.
Segment dynamic areas, and supplement the multi-frame cues with monocular cues. Limitations: • Uncontrolled segmentation quality;. • Additional segmentation ...
Many multi-frame methods handle dynamic areas by identifying them with explicit masks and compensating the multi-view cues with monocular cues represented as ...
People also ask
The representative approach [19,50] is to supervise dynamic areas of multi-frame depth with singleframe depth by using an additional training loss term, aiming ...
Improving Monocular Visual Odometry Using Learned Depth ... Learning to Fuse Monocular and Multi-view Cues for Multi-frame Depth Estimation in Dynamic Scenes.
[CVPR 2023] Multi-frame depth estimation in dynamic scenes. -- Li, Rui, et al. "Learning to Fuse Monocular and Multi-view Cues for Multi-frame Depth Estimation ...