4RC: 4D Reconstruction via Conditional Querying Anytime and Anywhere

Highly accurate and efficient VSLAM system for Python

Official repository for the ICCV2023 paper "Kick Back & Relax: Learning to Reconstruct the World by Watching SlowTV"

A Plugin-Based Multi-Agent System for In-Editor Academic Writing, Review, and Editing

Code release for "UnSAMv2: Self-Supervised Learning Enables Segment Anything at Any Granularity"

SAM 3D Objects

[NeurIPS 2025 Spotlight] "SANSA: Unleashing the Hidden Semantics in SAM2 for Few-Shot Segmentation."

Removing Cost Volumes from Optical Flow Estimators (ICCV 2025 Oral)

[ICCV 2025 Oral] Back on Track: Bundle Adjustment for Dynamic Scene Reconstruction (BA-Track)

[ICLR '26 Oral] Official repository of the paper "AnyUp: Universal Feature Upsampling".

Midway Network: Learning Representations for Recognition and Motion from Latent Dynamics

MapAnything: Universal Feed-Forward Metric 3D Reconstruction

PooDLe: Pooled and dense self-supervised learning from naturalistic videos

[ECCV 2024] Improving 2D Feature Representations by 3D-Aware Fine-Tuning

Motion-Refined DINOSAUR for Unsupervised Multi-Object Discovery (ICCVW 2025)

Activation Subspaces for Out-of-Distribution Detection (ICCV 2025)

Reference PyTorch implementation and models for DINOv3

Official Implementation of Marginalized Generalized IOU (MGIoU)

Scene-Centric Unsupervised Panoptic Segmentation (CVPR 2025 Highlight)

ART: Adaptive Relation Tuning for Generalized Relation Detection (ICCV 2025)

[ICLR2026] Official Implementation of "Dens3R: A Foundation Model for 3D Geometry Prediction"

[ICLR2025] A PyTorch implementation for STORM: Spatiotemporal Reconstruction Model for Large-Scale Outdoor Scenes

Feed-Forward SceneDINO for Unsupervised Semantic Scene Completion (ICCV 2025)

AllTracker is a model for tracking all pixels in a video.

[CVPR 2025 Highlight] Official code and models for Encoder-only Mask Transformer (EoMT).

GLASS: Guided Latent Slot Diffusion for Object-Centric Learning (CVPR 2025)