High-efficiency LLM inference engine in C++/CUDA. Run Llama 70B on RTX 3090.

Multimodal object tracking and scene analytics for highly actionable, real-world contextualized data

DINO, SAM, VJEPA and BLIP object detection, tracking, scene analysis and captioning test suite

A Unity Editor extension that provides a personal cross-project asset library, allowing developers to quickly import frequently used assets without manual folder navigation.

Vociferous captures speech, transcribes it locally with faster-whisper, and refines it with local or optional external models. Open-source, local-first, and built for private speech records.

Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery

Open-source Expressive Robot Lamp

[ICCV 2025] SpatialTrackerV2: 3D Point Tracking Made Easy

[ICRA 2025] MAC-VO: Metrics-aware Covariance for Learning-based Stereo Visual Odometry

[CVPR 2025 Best Paper Nomination] FoundationStereo: Zero-Shot Stereo Matching

Official repository for "AM-RADIO: Reduce All Domains Into One"

[CVPR'24 Highlight & Best Demo Award] Gaussian Splatting SLAM

A C++20 library for fast serialization, deserialization and validation using reflection. Supports JSON, Avro, Boost Serialization, BSON, Cap'n Proto, CBOR, Cereal, CSV, flexbuffers, msgpack, parque…

Shows a Service Framework Server to help a Magic Leap 2 user update to new versions op an app

Project for paper "Learning 3D Human Dynamics from Video"

[ICCV21] Self-Calibrating Neural Radiance Fields

[NeurIPS2022] Official implementation of PeRFception: Perception using Radiance Fields.

[ECCV22] BungeeNeRF: Progressive Neural Radiance Field for Extreme Multi-scale Scene Rendering

⚡ Based on yolo's ultra-lightweight universal target detection algorithm, the calculation amount is only 250mflops, the ncnn model size is only 666kb, the Raspberry Pi 3b can run up to 15fps+, and …

FAIR's research platform for object detection research, implementing popular algorithms like Mask R-CNN and RetinaNet.

Models and examples built with TensorFlow

Code repository for the paper: Hierarchical Kinematic Probability Distributions for 3D Human Shape and Pose Estimation from Images in the Wild (ICCV 2021)

Super Fast and Accurate 3D Object Detection based on 3D LiDAR Point Clouds (The PyTorch implementation)

Super Fast and Accurate 3D Object Detection based on 3D LiDAR Point Clouds (The PyTorch implementation)

Python Radio Over IP Module

Implementation of TimeSformer from Facebook AI, a pure attention-based solution for video classification