Computer vision

Official code for CVPR 2026 paper: VGGT-Det: Mining VGGT Internal Priors for Sensor-Geometry-Free Multi-View Indoor 3D Object Detection

A C++ library for ground segmentation of a pointcloud

Local AI photo scoring, culling, and gallery — score, organise, and explore your library with face recognition and semantic search. No cloud, no subscriptions.

Official repository of Utonia: Toward One Encoder for All Point Clouds

Privacy-first business card scanner that extracts structured contacts from photos/videos using configurable AI providers or on-device OCR, with review/edit and Excel/CSV/vCard export

Real-time AI assistant for Meta Ray-Ban smart glasses -- voice + vision + agentic actions via Gemini Live and OpenClaw

Official codebase for "Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation"

Visual Causal Flow

SentinelSight- AI Video Analytics

Youtu-Parsing: Perception, Structuring and Recognition via High-Parallelism Decoding

Official implementation of "VideoMaMa: Mask-Guided Video Matting via Generative Prior", CVPR 2026

Edit Banana: A framework for converting statistical formats into editable.

Code for the ShapeR research paper

VerseCrafter: Dynamic Realistic Video World Model with 4D Geometric Control

Towards Pixel-Level VLM Perception via Simple Points Prediction

[TPAMI 2026] Interacted Planes Reveal 3D Line Mapping

Create a digital-twin style traffic visualization using only mp4 CCTV footage and its Google Maps location.

一款基于 Tauri 框架开发的现代化 Windows 面容识别解锁增强软件。它通过自定义 Credential Provider (DLL) 注入 Windows 登录界面，结合前端 Vue 3 和后端 OpenCV 人脸识别算法，为用户提供类似 Windows Hello 的解锁体验。

Official Codebase for our CVPR 2026 paper UniSH: Unifying Scene and Human Reconstruction in a Feed-Forward Pass

An open source library designed to provide community examples of Joint Embedding Predictive Architectures (JEPAs). It contains code and examples for learning representations from images, video, and action-conditioned video, as well as planning using JEPA-based models.