"E-RayZer: Self-supervised 3D Reconstruction as Spatial Visual Pre-training" official implementation.

📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.🎉

A differentiable rasterizer used in the project "2D Gaussian Splatting"

The code for the paper "Editable Physically-based Reflections in Raytraced Gaussian Radiance Fields"

Depth Anything 3

Contexts Optical Compression

Public code for XFactor: Introduces the first geometry-free model to achieve true self-supervised / pose-free Novel View Synthesis (NVS) by learning transferable latent camera pose representations.

Code for ICCV'2025 (Best student paper honorable mention) "RayZer: A Self-supervised Large View Synthesis Model"

[NeurIPS 2025] Pixel-Perfect Depth

A simple state update rule to enhance length generalization for CUT3R

MapAnything: Universal Feed-Forward Metric 3D Reconstruction

Code release for paper "Test-Time Training Done Right"

[ICCV 2025 Oral] MVTracker: Multi-view 3D Point Tracking

minLoRA: a minimal PyTorch library that allows you to apply LoRA to any PyTorch model.

Reference PyTorch implementation and models for DINOv3

ViPE: Video Pose Engine for Geometric 3D Perception

Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

Code of π^3: Permutation-Equivariant Visual Geometry Learning

[CVPR 2025] AerialMegaDepth: Learning Aerial-Ground Reconstruction and View Synthesis

VGGT-SLAM: Dense RGB SLAM Optimized on the SL(4) Manifold

Cameras as Relative Positional Encoding

JAX implementation of MeanFlow

Official implementation of our ICCV'25 paper "Learning Dense Feature Matching via Lifting Single 2D Image to 3D Space"

[CVPR 2024] LEAP-VO: Long-term Effective Any Point Tracking for Visual Odometry

run DROID-SLAM with Metric3D to improve monocular performance