#computer-vision Repositories
1,451 repositories total
Collection of common code that's shared among different research projects in FAIR computer vision team.
An open source library and framework for deep learning on satellite and aerial imagery.
A reasoning framework with many useful demos at the same time, if you find it useful, please give it a star
Labeling tool utilizing SAM (Segment Anything Model), supporting SAM, SAM2, SAM3, sam-hq, MobileSAM, EdgeSAM, etc. Interactive semi-automatic image annotation tool.
Production-grade C++ edge AI engine for video analytics and on-device VLM across Sophon, Rockchip RKNN, and x86, with visual orchestration, real-time OSD, events, and reproducible benchmarks.
This is a yolo3-pytorch source code, which can be used to train your own model.
This resource supports the author's Python image processing articles on CSDN, mainly focusing on the implementation of algorithms for image processing, image recognition, and image classification in Python. We hope this resource is helpful to you, and let's work hard together.
Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding
NVIDIA Alpamayo 1 Nano is an open 10B reasoning VLA model for autonomous vehicles that pairs driving trajectories with Chain-of-Causation reasoning.
Collaborate & label any type of data, images, text, or documents, in an easy web interface or desktop app.
Continuous Thought Machines, because thought takes time and reasoning is a process.
Everything about federated learning, including research papers, books, codes, tutorials, videos and beyond
https://textbehindimage.rexanwong.xyz - create text behind image designs easily
Comprehensive and timely academic information on federated learning (papers, frameworks, datasets, tutorials, workshops)
Python wrapper to Philipp Krรคhenbรผhl's dense (fully connected) CRFs with gaussian edge potentials.
Recipes for shrinking, optimizing, customizing cutting edge vision models. ๐
NVIDIA AI Blueprint for video search and summarization (VSS) is a GPU-accelerated reference architecture for building video analytics agents with real-time verified alerts, visual Q&A, and automated reporting. The VSS Blueprint uses vision language models (VLMs) such as NVIDIA Cosmos, LLMs such as NVIDIA Nemotron, RAG, and NVIDIA NIMs.
Worldwide building footprints derived from satellite imagery
Code release for "Detecting Twenty-thousand Classes using Image-level Supervision".
Approaching (Almost) Any Machine Learning Problem Chinese Version, Online Documentation Address: https://ytzfhqs.github.io/AAAMLP-CN/
Custom Diffusion: Multi-Concept Customization of Text-to-Image Diffusion (CVPR 2023)