We repurpose video diffusion models to reconstruct 4D geometry (4D pointmaps) from monocular video.
Mar 16, 2025
We introduce global bundle adjustment for NeRFs, significantly improving the quality of novel view synthesis from sparse views with noisy poses or even without poses.
Sep 1, 2024
We propose multiview video diffusion models that generate high-quality multi-view videos for dynamic objects from text.
Jun 14, 2024
An improvement of 3D Gaussian Splatting.
Jun 1, 2024
We propose to use both 2D and 3D diffusion models through SDS optimization to generate high-quality, textured 3D meshes from a single image in the wild.
Mar 14, 2024
We propose EgoLoc, a novel approach for localizing 3D objects from egocentric videos, achieving SOTA performance on the Ego4D VQ3D task with 87.12% success rate over previous SOTA of only 8%.
Oct 1, 2023
We revisit PointNet++ and propose PointNeXt, a next-generation PointNet architecture.
Nov 1, 2022
We propose EIL, a novel adversarial erasing technique for weakly supervised object localization.
Oct 1, 2020