keywords: DeepLearning, Tools, Frameworks, Examples & Papers



Nonrigid image registration using multi-scale 3D convolutional neural networks

Codebase for Image Classification Research, written in PyTorch.

Pytorch implementation of network design paradigm described in the paper “Designing Network Design Spaces”

Facebook AI RegNet Models Outperform EfficientNet Models, Run 5x Faster on GPUs



Tensors and Dynamic neural networks in Python with strong GPU acceleration.

Fine-tune pretrained Convolutional Neural Networks with PyTorch


Colossal-AI: A Unified Deep Learning System for Large-Scale Parallel Training


Comprehensive Library

Open Source Computer Vision Library

A collection of various deep learning architectures, models, and tips

Machine learning, in numpy

Virginia Tech Vision and Learning Lab

scikit-learn: machine learning in Python

Cross-platform, customizable ML solutions for live and streaming media. Face Detection, Face Mesh, Iris, Hands, Pose, Holistic.

Management Tools

Fastest unstructured dataset management for TensorFlow/PyTorch. Stream data real-time & version-control it.

This repository includes optimized deep learning models and a set of demos to expedite development of high-performance deep learning inference applications. Use these free pre-trained models instead of training your own models to speed-up the development and production deployment process.

This toolkit allows developers to deploy pre-trained deep learning models through a high-level C++ Inference Engine API integrated with application logic.

Recommender System

An open source recommender system service written in Go.

Shape Representation

Learning Continuous Signed Distance Functions for Shape Representation

This repository contains the code for the paper “Occupancy Networks - Learning 3D Reconstruction in Function Space”

3D Data

PyTorch3D is FAIR’s library of reusable components for deep learning with 3D data

Object Detection

Detectron2 is FAIR’s next-generation research platform for object detection and segmentation.

This project provides the implementation for DetNAS: Backbone Search for Object Detection.

Face Recognition

Deepfakes Software For All

An open source library for face detection in images. The face detection speed can reach 1500FPS.

DeepFaceLab is the leading software for creating deepfakes.

Image Recognition

Deep learning model for recognizing puzzle patterns in The Witness.
I trained a robot to play The Witness
Is a Realistic Honey Simulation Possible?

Fake Detector

FaceForensics++: Learning to Detect Manipulated Facial Images

CenterFace(size of 7.3MB) is a practical anchor-free face detection and alignment method for edge devices.

Image Augmentation

Image augmentation for machine learning experiments.

Depth-Aware video frame INterpolation (DAIN)

Depth-Aware Video Frame Interpolation (CVPR 2019)

Gesture Recognizer

Gesture recognition via CNN. Implemented in Keras + Tensorflow/Theano + OpenCV

基于卷积神经网络的数字手势识别安卓APP,识别数字手势0-10(The number gestures recognition Android APP based on convolutional neural network(CNN), which can recognize the gestures corresponding number 0 to 10)

Real-time Hand Gesture Recognition with PyTorch on EgoGesture, NvGesture and Jester

Image & Video Classification

An end-to-end PyTorch framework for image and video classification

Turi Create simplifies the development of custom machine learning models.

Image Generataion

Progressive Growing of GANs for Improved Quality, Stability, and Variation


A small course on exploiting and defending neural networks

AI Path Tracer Denoiser

This project is to create a CUDA accelerated Deep learning approach to denoise renders from a path tracer


[CVPR 2020] 3D Photography using Context-aware Layered Depth Inpainting

NVIDIA Merlin is a framework for building high-performance, deep learning-based recommender systems.
Announcing NVIDIA Merlin: An Application Framework for Deep Recommender Systems

Pose Estimation

COCO-WholeBody dataset is the first large-scale benchmark for whole-body pose estimation.

GCN (Graph Convolutional Networks)

Graph Convolutional Networks in PyTorch

Graph Convolution Network for PyTorch

Hierarchical Task Network (HTN)

Hierarchical task network

Data Loading

The NVIDIA Data Loading Library (DALI) is a GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.


Shape Representation

DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation

Occupancy Networks: Learning 3D Reconstruction in Function Space

Physical Animation

Try W&B to track machine learning experiments and visualize results
AI Learns To Compute Game Physics In Microseconds

Hardware Architecture

Large-scale Deep Unsupervised Learning using Graphics Processors

Fast Support Vector Machine Training and Classification on Graphics Processors

HeteroSpark: A Heterogeneous CPU/GPU Spark Platform for Machine Learning Algorithms


Self-learning Transformations for Improving Gaze and Head Redirection

3D Scenes

Pose2Room: Understanding 3D Scenes from Human Activities

Racing Game

Game AI: Simulating Car Racing Game by Applying Pathfinding Algorithms

Reinforcement Learning for a Simple Racing Game



Research in human-centered AI, deep learning, autonomous vehicles & robotics at MIT and beyond.


Using neural networks to solve advanced mathematics equations


Youtube Tutorials

Neural Network Architectures

MindsDB is a predictive AI layer for existing databases, making it easy for organizations to apply machine learning to their own data.
Machine Learning in one line of code

OpenAI Plays Hide and Seek…and Breaks The Game!



Deep Learning for Coders with fastai and PyTorch: AI Applications Without a PhD 1st Edition (July 14, 2020)

Deep Learning with PyTorch 1st Edition (June 9, 2020)

Commercial Product

Code Review

Write better code with the knowledge of the global development community.


Instantly Transform Any Text Into A 100% Human-Sounding Voiceover with only 3 clicks!


An A.I. agent for visual tasks

Digital Creative

Runway is a new kind of creative suite. One where AI is a collaborator and anything you can imagine can be created.

Stable Diffusion is a deep learning, text-to-image model released in 2022.


AI Completions. Never Code Alone.

Social Accounts

Twitter Accounts

Research Scientist at @GoogleAI in the Brain Team. Deep Learning with Graphs.

maths, visualisations, conversational AI. lead scientist @Poly_AI, previously @GoogleAI, PhD @Cambridge_Uni. living in Singapore.


Dance Motion

This repo contains starter code for using the AIST++ dataset.

It was inevitable: the scent of bitter almonds always reminded him of the fate of unrequited love. ― Gabriel Garcia Marquez