Trending Projects

A 3D object proposal generation and detection library from point cloud data, useful for computer vision.

A survey paper that proposes a taxonomy of Retrieval-Augmented Generation (RAG) techniques for AI-Generated Content (AIGC).

PyTorch-based library for image classification tasks like CIFAR-10, CIFAR-100, and ImageNet.

A wrapper library around the ChatGPT and OpenAI HTTP API, enabling Swift developers to integrate AI tools into their apps.

A deep learning library for head pose estimation using PyTorch, useful for computer vision applications.

A fast and scalable SVM library for classification and regression tasks on GPUs and CPUs.

A survey paper on evaluating large language models (LLMs) for developers building AI-powered applications.

A repository for training large language models (LLMs) using QLoRA and FSDP techniques.

An IBM toolkit for computing on encrypted data without decryption, with demos for secure ML and key-value search.

An AI-powered search engine for developers to discover tools and resources for building with AI.

This open-source repository provides a comprehensive set of Chinese tutorials for the popular OpenCV computer vision library in Python.

A highly accurate deep learning-based license plate detection and recognition system for real-time use.

A PyTorch library for semantic segmentation, a computer vision task to classify each pixel in an image.

Demonstrations of Magenta Models, a set of open-source machine learning models for music and art generation.

A reactive runtime for building durable AI agents using Rust, with a focus on LLM orchestration and debugging.

Minimal Deep Q Learning (DQN & DDQN) implementations in Python using the Keras library for reinforcement learning.

A curated collection of resources on applying Transformers to medical imaging tasks like segmentation, classification, and synthesis.

An open-source chatbot framework for developers to build AI-powered chat applications.

VideoLLaMA 2 is a Python library that advances spatial-temporal modeling and audio understanding in video-based large language models.

Convolutional neural network for analyzing pentest screenshots, useful for security researchers and penetration testers.

Distributed deep learning on Hadoop and Spark clusters for vibe coders.

A Python library for building Bayesian statistical models and performing Bayesian inference.

Official implementation of a PyTorch-based library for semantic keypoint detection and pose estimation.

A central hub for extending OpenMMLab's powerful computer vision capabilities with SAM and other AI features.

A curated list of resources for relation extraction, a key NLP task, useful for vibe coders.

Adaptive is a parallel active learning library for efficiently learning mathematical functions.

A template for deploying a Keras-based image classifier web app using Flask in 10 minutes.

A high-performance entity linking solution for linking mentions to their corresponding entities in text.

A high-capacity image restoration model focused on privacy-safe dataset curation for AI-powered tooling.

Calamari-OCR/calamari is a line-based ATR engine based on OCRopy, written in Python.

This Jupyter Notebook series covers the fundamentals of NLP and Computer Vision, leading to cutting-edge Vision-Language Models.

A point-based neural radiance field for 3D reconstruction and rendering from multi-view images.

A Java library for developers building AI-powered applications and tools.

A Keras implementation of image outpainting, allowing developers to extend images beyond their borders.

Open-source audio annotation tool for machine learning and speech processing datasets.

A tool for structurally pruning large language models like LLaMA, BLOOM, and Vicuna to reduce their size and inference time.

A generic C# GOAP (Goal Oriented Action Planning) library with Unity3d examples for game developers.

A Python library that generates 3D textured meshes from text prompts using 2D text-to-image models.

A research project focused on open-world object detection using continual and contrastive learning techniques.

limdu is a machine learning library for Node.js that provides a flexible and extensible framework for building AI applications.

A TensorFlow implementation of Neural Turing Machines, a class of models that can learn algorithms.

Panda is an open-source large language model project for the Chinese language, aiming to drive innovation and collaboration in NLP.

A Python library for converting the Llama language model to ONNX format for faster inference.

Next.js frontend for the LangChain conversational AI framework, enabling developers to build AI-powered chat applications.

A series of large language models trained from scratch by developers @01-ai for vibe coders.

Plugin for Stable Diffusion WebUI that optimizes large image generation through tiled diffusion and VAE techniques.

Run Keras models in the browser, with GPU support using WebGL

Infrastructure components for cost-efficient GenAI model inference and deployment

A personal GitHub page for a computer vision and deep learning researcher.

A Python-based OCR library that uses CTPN, CRNN, and CTC to perform text detection and recognition in natural scenes.