Comprehensive, latest, and deployable video deep learning algorithm, including video recognition, action localization, and temporal action detection tasks. It's a high-performance, light-weight cod…

SAM: Sharpness-Aware Minimization (PyTorch)

PySlowFast: video understanding codebase from FAIR for reproducing state-of-the-art video models.

Pre-trained and Reproduced Deep Learning Models （『飞桨』官方模型库，包含多种学术前沿和工业场景验证的深度学习模型）

Reference models and tools for Cloud TPUs.

[ICCV 2019] TSM: Temporal Shift Module for Efficient Video Understanding

A PyTorch implementation of EfficientNet

This repository reproduces the results of the paper: "Fixing the train-test resolution discrepancy" https://arxiv.org/abs/1906.06423

Starter code for working with the YouTube-8M dataset.

Code of the paper "SplitNet: Divide and Co-training".

detector for rotated-objections based on CenterNet/基于CenterNet的旋转目标检测

Sharp Image Deblurring

Lifespan Age Transformation Synthesis code

Unofficial implemention of lanenet model for real time lane detection using deep neural network model https://maybeshewill-cv.github.io/lanenet-lane-detection/

Image Super-Resolution for Anime-Style Art

(ImageNet pretrained models) The official pytorch implemention of the TPAMI paper "Res2Net: A New Multi-scale Backbone Architecture"

YOLOv4 - Neural Networks for Object Detection (Windows and Linux version of Darknet )

Image augmentation for machine learning experiments.

Federated Learning Library: https://fedml.ai

waifu2x converter ncnn version, runs fast on intel / amd / nvidia GPU with vulkan

Federated Learning - PyTorch

Retinaface get 80.99% in widerface hard val using mobilenet0.25.

The most influential papers in convolutional attention mechanisms, suitable for image classification, image and video object segmentation and tracking

Source Codes for "Human-object Relation Network for Action Recogntion in Still Images"

⚡ Yolo universal target detection model combined with EfficientNet-lite, the calculation amount is only 230Mflops(0.23Bflops), and the model size is 1.3MB

Keras implementation of a CNN network for age and gender estimation

Neural Blind Deconvolution Using Deep Priors (CVPR 2020)

MobileNetV2-YoloV3-Nano: 0.5BFlops 3MB HUAWEI P40: 6ms/img, YoloFace-500k:0.1Bflops 420KB:fire::fire::fire: