PURE: Refining Feed-forward 3D Reconstruction with Unlabeled Post-training
cs.CV
Submitted: 2026-10-10
Updated: 2026-10-10
Sources
- Virtual KITTI 2
- Reconstructing 4D Spatial Intelligence: A Survey
- LEAP-VO: Long-term Effective Any Point Tracking for Visual Odometry
- TTT3R: 3D Reconstruction as Test-Time Training
- Diffusion as Shader: 3D-aware Video Diffusion for Versatile Video Generation Control
- ViPE: Video Pose Engine for 3D Geometric Perception
- PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation
- Pow3R: Empowering Unconstrained 3D Reconstruction with Camera and Scene Priors
- T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy
- Segment Anything
- UniSH: Unifying Scene and Human Reconstruction in a Feed-Forward Pass
- Mono3R: Exploiting Monocular Cues for Geometric 3D Reconstruction
- Sekai: A Video Dataset towards World Exploration
- Depth Anything 3: Recovering the Visual Space from Any Views
- MoDGS: Dynamic Gaussian Splatting from Casually-captured Monocular Videos with Depth Priors
- Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection
- Align3R: Aligned Monocular Depth Estimation for Dynamic Videos
- TrackingWorld: World-centric Monocular 3D Tracking of Almost All Pixels
- ReFusion: 3D Reconstruction in Dynamic Environments for RGB-D Cameras Exploiting Residuals
- SAM 2: Segment Anything in Images and Videos
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models