Abstract: Visual odometry (VO) is the core part of visual simultaneous localization and mapping (SLAM). Most of the existing learning-based VO methods adopt either a framework for processing adjacent ...
a novel visual odometry approach that robustly handles regions violating the static scene assumption within an unsupervised visual odometry framework. python ...
Abstract: We propose MAC-VO, a novel learning-based stereo visual odometry (VO) framework that trains a metrics-aware uncertainty model to serve two critical functions: selecting keypoints and ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results