GRC-Pose is introduced, a correspondence-based framework that combines learned correspondence prediction with robust pose estimation and achieves state-of-the-art Average Recall and motion retention on HOT3D, improving the latter by 58% over prior art.
Experiments show that COVER improves the coverage-conflict trade-off, making CM-EVS a sparse, compact, and auditable RGB-D-pose resource for geometry-consistent panoramic 3D learning.
VDGS introduces visibility-driven statistics for scene anchors to quantify supervision strength and is leveraged for scene partitioning and for gradient compensation in under-optimized regions, thereby promoting balanced optimization across different regions.
Hao-Lin Yu, Jia-Dong Tang, Yi-Xian Wang et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.