Click here to flash read.
arXiv:2404.15263v1 Announce Type: new
Abstract: We introduce a new system for Multi-Session SLAM, which tracks camera motion across multiple disjoint videos under a single global reference. Our approach couples the prediction of optical flow with solver layers to estimate camera pose. The backbone is trained end-to-end using a novel differentiable solver for wide-baseline two-view pose. The full system can connect disjoint sequences, perform visual odometry, and global optimization. Compared to existing approaches, our design is accurate and robust to catastrophic failures. Code is available at github.com/princeton-vl/MultiSlam_DiffPose
Click here to read this post out
ID: 820106; Unique Viewers: 0
Unique Voters: 0
Total Votes: 0
Votes:
Latest Change: April 24, 2024, 7:31 a.m.
Changes:
Dictionaries:
Words:
Spaces:
Views: 10
CC:
No creative common's license
No creative common's license
Comments: