Hang Zhang

Kimi-K2.5. Moonshot's most powerful model

Kimi-K2. Kimi K2 is the large language model series developed by Moonshot AI team

vla_tutorial. HTML

Emu3.5. Native Multimodal Models are World Learners

lightning-attention. Lightning Attention-2: A Free Lunch for Handling Unlimited Sequence Lengths in Large Language Models

Liger-Kernel. Efficient Triton Kernels for LLM Training

Muon. Muon is an optimizer for hidden layers in neural networks

open-r1. Fully open reproduction of DeepSeek-R1

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

awesome-deepseek-integration. Integrate the DeepSeek API into popular software

DeepSeek-R1.

DeepSeek-VL2. DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding

Janus. Janus-Series: Unified Multimodal Understanding and Generation Models

mar. PyTorch implementation of MAR+DiffLoss https://arxiv.org/abs/2406.11838

Awesome-LLM4AD. A curated list of awesome LLM/VLM/VLA/World Model for Autonomous Driving(LLM4AD) resources (continually updated)

nanoGPT. The simplest, fastest repository for training/finetuning medium-sized GPTs.

LivePortrait. Bring portraits to life!

airflow. Apache Airflow - A platform to programmatically author, schedule, and monitor workflows

omni3d. Code release for "Omni3D A Large Benchmark and Model for 3D Object Detection in the Wild"

ViT-Adapter. [ICLR 2023 Spotlight] Vision Transformer Adapter for Dense Predictions

temporal-shift-module. [ICCV 2019] TSM: Temporal Shift Module for Efficient Video Understanding

flash-attention. Fast and memory-efficient exact attention

EMM-for-stock-prediction. We propose a model to analyze sentiment of online stock forum and use the information to predict stock volatility in the Chinese market. By generating a sentimental dictionary, we analyze the sentimental tendencies of each post as sentiment indicators. Such sentimental information will be fused with market data for prediction based on Recurrent Neural Networks (RNNs). We manually labeled the sentiment of forum post and make the data public available for research. Empirical evidence shows that 8 of the 10 stocks perform better with sentimental indicators.

Auto_painter. Recently, realistic image generation using deep neural networks has become a hot topic in machine learning and computer vision. Such an image can be generated at pixel level by learning from a large collection of images. Learning to generate colorful cartoon images from black-and-white sketches is not only an interesting research problem, but also a useful application in digital entertainment. In this paper, we investigate the sketch-to-image synthesis problem by using conditional generative adversarial networks (cGAN). We propose a model called auto-painter which can automatically generate compatible colors given a sketch. Wasserstein distance is used in training cGAN to overcome model collapse and enable the model converged much better. The new model is not only capable of painting hand-draw sketch with compatible colors, but also allowing users to indicate preferred colors. Experimental results on different sketch datasets show that the auto-painter performs better than other existing image-to-image methods.

CoupleGenerator. Generate your lover with your photo

TorchDistiller. Python

ETC-Real-time-Per-frame-Semantic-video-segmentation. Enforcing temporal consistency in real-time per-frame semantic video segmentation

structure_knowledge_distillation. The official code for the paper 'Structured Knowledge Distillation for Semantic Segmentation'. (CVPR 2019 ORAL) and extension to other tasks.

DriveLM. [ECCV 2024 Oral] DriveLM: Driving with Graph Visual Question Answering

review_object_detection_metrics. Object Detection Metrics. 14 object detection metrics: mean Average Precision (mAP), Average Recall (AR), Spatio-Temporal Tube Average Precision (STT-AP). This project supports different bounding box formats as in COCO, PASCAL, Imagenet, etc.

SST. Code for a series of work in LiDAR perception, including SST (CVPR 22), FSD (NeurIPS 22), FSD++ (TPAMI 23), FSDv2, and CTRL (ICCV 23, oral).

StreamPETR. [ICCV 2023] StreamPETR: Exploring Object-Centric Temporal Modeling for Efficient Multi-View 3D Object Detection

BEVFormer. [ECCV 2022] This is the official implementation of BEVFormer, a camera-only framework for autonomous driving perception, e.g., 3D object detection and semantic map segmentation.

UniAD. [CVPR 2023 Best Paper Award] Planning-oriented Autonomous Driving

ncnn. ncnn is a high-performance neural network inference framework optimized for the mobile platform

PETR. [ECCV2022] PETR: Position Embedding Transformation for Multi-View 3D Object Detection & [ICCV2023] PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images

GroundingDINO. [ECCV 2024] Official implementation of the paper "Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection"

adaptive_teacher. This repo provides the source code for "Cross-Domain Adaptive Teacher for Object Detection".

unbiased-teacher-v2. PyTorch code for CVPR 2022 paper Unbiased Teacher v2 Semi-supervised Object Detection for Anchor-free and Anchor-based Detectors

FBGEMM. FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/

model-quantization. Collections of model quantization algorithms. Any issues, please contact Peng Chen (blueardour@gmail.com)

ov-seg. This is the official PyTorch implementation of the paper Open-Vocabulary Semantic Segmentation with Mask-adapted CLIP.

bevfusion. [ICRA'23] BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation

frustum-convnet. The PyTorch Implementation of F-ConvNet for 3D Object Detection

Awesome-Monocular-3D-Detection. Awesome Monocular 3D detection

GUPNet. Python

dd3d. Official PyTorch implementation of DD3D: Is Pseudo-Lidar needed for Monocular 3D Object detection? (ICCV 2021), Dennis Park*, Rares Ambrus*, Vitor Guizilini, Jie Li, and Adrien Gaidon.

detrex. detrex is a research platform for DETR-based object detection, segmentation, pose estimation and other visual recognition tasks.

3detr. Code & Models for 3DETR - an End-to-end transformer model for 3D object detection

frustum-pointnets. Frustum PointNets for 3D Object Detection from RGB-D Data

MaskDINO. [CVPR 2023] Official implementation of the paper "Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation"

zsseg.baseline. Open-vocabulary Semantic Segmentation

ConvMAE. ConvMAE: Masked Convolution Meets Masked Autoencoders

metaseq. Repo for external large-scale work

OTTER. This code provides a PyTorch implementation for OTTER (Optimal Transport distillation for Efficient zero-shot Recognition), as described in the paper.

CLIP. CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

nerf-pytorch. A PyTorch implementation of NeRF (Neural Radiance Fields) that reproduces the results.

volrend. PlenOctree Volume Rendering (supports CUDA & fragment shader backends)

TensoRF. [ECCV 2022] Tensorial Radiance Fields, a novel approach to model and reconstruct radiance fields

detectron2. Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.

BladeDISC. BladeDISC is an end-to-end DynamIc Shape Compiler project for machine learning workloads.

DAB-DETR. [ICLR 2022] Official implementation of the paper "DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETR"

CVPR-2021-Paper-Statistics. Statistics and Visualization of acceptance rate, main keyword of CVPR 2021 accepted papers for the main Computer Vision conference (CVPR)

awesome-detection-transformer. Collect some papers about transformer for detection and segmentation. Awesome Detection Transformer for Computer Vision (CV)

DINO. [ICLR 2023] Official implementation of the paper "DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection"

DN-DETR. [CVPR 2022 Oral] Official implementation of DN-DETR

Ax. Adaptive Experimentation Platform

nanodet. NanoDet-Plus⚡Super fast and lightweight anchor-free object detection model. 🔥Only 980 KB(int8) / 1.8MB (fp16) and run 97FPS on cellphone🔥

MAE-pytorch. Unofficial PyTorch implementation of Masked Autoencoders Are Scalable Vision Learners

AlphaNet. AlphaNet Improved Training of Supernet with Alpha-Divergence

Mask2Former. Code release for "Masked-attention Mask Transformer for Universal Image Segmentation"

K-Net. [NeurIPS2021] Code Release of K-Net: Towards Unified Image Segmentation

wetectron. Weakly-supervised object detection.

XUAN.

Dummy-Robot. 我的超迷你机械臂机器人项目。

AnchorDETR. An official implementation of the Anchor DETR.

centerX. This repo is implemented based on detectron2 and centernet

CoTNet. This is an official implementation for "Contextual Transformer Networks for Visual Recognition".

FaPN. [ICCV 2021] FaPN: Feature-aligned Pyramid Network for Dense Image Prediction

unilm. Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

ConditionalDETR. This repository is an official implementation of the ICCV 2021 paper "Conditional DETR for Fast Training Convergence". (https://arxiv.org/abs/2108.06152)

AdelaiDet. AdelaiDet is an open source toolbox for multiple instance-level detection and recognition tasks.

SOTR. SOTR: Segmenting Objects with Transformers

Yolact_fcos. YOLACT: Real-time Instance Segmentation on the FCOS detector (without bbox cropping), achives 35.2mAP on coco val

ReSim. PyTorch Implementation of Region Similarity Representation Learning (ReSim)

large-vocab-devil. Python

detectron2_backbone. detectron2 backbone: resnet18, efficientnet, hrnet, mobilenet v2, resnest, bifpn

MaskFormer. Per-Pixel Classification is Not All You Need for Semantic Segmentation (NeurIPS 2021, spotlight)

hydra. Hydra is a framework for elegantly configuring complex applications

slidev. Presentation Slides for Developers

dino. PyTorch code for Vision Transformers training with the Self-Supervised learning method DINO

SelfSup. Collections of self-supervised methods, based on cvpods.

buck. A fast build system that encourages the creation of small, reusable modules over a variety of platforms and languages.

openrazer. Open source driver and user-space daemon to control Razer lighting and other features on GNU/Linux

razer-macos. Color effects manager for Razer devices for macOS. Supports High Sierra (10.13) to Monterey (12.0). Made by the community, based on openrazer.

pit. Python

vega. AutoML tools chain

torchtyping. Type annotations and dynamic checking for a tensor's shape, dtype, names, etc.

mmdetection3d. OpenMMLab's next-generation platform for general 3D object detection.

do-you-even-need-attention. Is the attention layer even necessary? (https://arxiv.org/abs/2105.02723)

Swin-Transformer. This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".

nni. An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.

nevergrad. A Python toolbox for performing gradient-free optimization

CenterNet2. Two-stage CenterNet

ptr. Python Test Runner.

black. The uncompromising Python code formatter

pytorch-classification-advprop. Python

d2go. D2Go is a toolkit for efficient deep learning

model_search. Python

unbiased-teacher. PyTorch code for ICLR 2021 paper Unbiased Teacher for Semi-Supervised Object Detection

deeplake. Deeplake is AI Data Runtime for Agents. It provides serverless postgres with a multimodal datalake, enabling scalable retrieval and training.

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

deepmind-research. This repository contains implementations and illustrative code to accompany DeepMind publications

DetCo. Python

googlefinance. Python module to get real-time stock data from Google Finance API

deit. Official DeiT repository

vissl. VISSL is FAIR's library of extensible, modular and scalable components for SOTA Self-Supervised Learning with images.

SparseR-CNN. [CVPR2021, PAMI2023] End-to-End Object Detection with Learnable Proposal

RepVGG. RepVGG: Making VGG-style ConvNets Great Again

DeFCN. End-to-End Object Detection with Fully Convolutional Network

yacs. YACS -- Yet Another Configuration System

pytorch-lightning. Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.

dynamic-training-with-apache-mxnet-on-aws. Dynamic training with Apache MXNet reduces cost and time for training deep neural networks by leveraging AWS cloud elasticity and scale. The system reduces training cost and time by dynamically updating the training cluster size during training, with minimal impact on model training accuracy.

awesome-transformer-for-vision.

vision_transformer. Jupyter Notebook

SETR. [CVPR 2021 & IJCV 2024] Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers

eagerpy. PyTorch, TensorFlow, JAX and NumPy — all of them natively using the same code

jax-resnet. Implementations and checkpoints for ResNet, Wide ResNet, ResNeXt, ResNet-D, and ResNeSt in JAX (Flax).

ContextPrior. Implementation for "Context Prior for Scene Segmentation"

irmae. PyTorch implementation of IRMAE https//arxiv.org/abs/2010.00679

Deformable-DETR. Deformable DETR: Deformable Transformers for End-to-End Object Detection.

micrograd. A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API

tinygrad. You like pytorch? You like micrograd? You love tinygrad! ❤️

pyre-check. Performant type-checking for python.

slurm. Slurm: A Highly Scalable Workload Manager

submitit. Python 3.8+ toolbox for submitting jobs to Slurm

slimmable_networks. Slimmable Networks, AutoSlim, and Beyond, ICLR 2019, and ICCV 2019

mobile-vision. Mobile vision models and code

Computer-Vision-Leaderboard. Comparison of famous convolutional neural network models

htm.core. Actively developed Hierarchical Temporal Memory (HTM) community fork (continuation) of NuPIC. Implementation for C++ and Python

WeightNet. Python

PyTorch_YOLOv4. PyTorch implementation of YOLOv4

AtomNAS. [ICLR 2020]: 'AtomNAS: Fine-Grained End-to-End Neural Architecture Search'

tensorflow-yolov4-tflite. YOLOv4, YOLOv4-tiny, YOLOv3, YOLOv3-tiny Implemented in Tensorflow 2.0, Android. Convert YOLO v4 .weights tensorflow, tensorrt and tflite

pytorch-material-classification. Material Classification with Convolutional Networks in PyTorch

pytorch_simple_CenterNet_45. A simple pytorch implementation of CenterNet (Objects as Points)

nas-without-training. Code for Neural Architecture Search without Training (ICML 2021)

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

openseg.pytorch. The official Pytorch implementation of OCNet, OCRNet, and SegFix.

SSD. High quality, fast, modular reference implementation of SSD in PyTorch

pytorch-YOLOv4. PyTorch ,ONNX and TensorRT implementation of YOLOv4

mmselfsup. OpenMMLab Self-Supervised Learning Toolbox and Benchmark

mmsegmentation. OpenMMLab Semantic Segmentation Toolbox and Benchmark.

awesome-lane-detection. A paper list of lane detection.

yolov3. Ultralytics YOLOv3 in PyTorch > ONNX > CoreML > TFLite

onnx_tflite_yolov3. A Conversion tool to convert YOLO v3 Darknet weights to TF Lite model (YOLO v3 PyTorch > ONNX > TensorFlow > TF Lite), and to TensorRT (YOLO v3 Pytorch > ONNX > TensorRT).

Auto-PyTorch. Automatic architecture search and hyperparameter optimization for PyTorch

detr. End-to-End Object Detection with Transformers

pytorch3d. PyTorch3D is FAIR's library of reusable components for deep learning with 3D data

PlotNeuralNet. Latex code for making neural networks diagrams

s3cmd. Official s3cmd repo -- Command line tool for managing S3 compatible storage services (including Amazon S3 and CloudFront).

yolov5. Ultralytics YOLOv5 in PyTorch > ONNX > CoreML > TFLite

PaddleClas. A treasure chest for visual classification and recognition powered by PaddlePaddle

ResNeSt-TF2. TensorFlow implementation of "ResNeSt: Split-Attention Networks"

hub. Submission to https://pytorch.org/hub/

once-for-all. [ICLR 2020] Once for All: Train One Network and Specialize it for Efficient Deployment

ResNeSt-caffe. A Caffe version of official PyTorch ResNeSt

DetectoRS. DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution

pycls. Codebase for Image Classification Research, written in PyTorch.

pytorch-image-models. The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

PyContrast. PyTorch implementation of Contrastive Learning methods

ENAS-pytorch. PyTorch implementation of "Efficient Neural Architecture Search via Parameters Sharing"

Fast-AutoAug-Torch. Python

RegNet-Search-PyTorch. Search for RegNet using PyTorch

imglab. To speedup and simplify image labeling/ annotation process with multiple supported formats.

tf_ResNeSt_RegNet_model. tensorflow 2.x version of ResNeSt,RegNet,DETR

BayesianOptimization. A Python implementation of global optimization with gaussian processes.

CenterNet. Object detection, 3D detection, and pose estimation using center point detection:

CenterTrack. Simultaneous object detection and tracking using center points.

pysot. SenseTime Research platform for single object tracking, implementing algorithms like SiamRPN and SiamMask.

pytorch-retinanet. Pytorch implementation of RetinaNet object detection.

retinanet-examples. Fast and accurate object detection with end-to-end GPU optimization

Face_Pytorch. face recognition algorithms in pytorch framework, including arcface, cosface, sphereface and so on

efficientdet-pytorch. A PyTorch impl of EfficientDet faithful to the original Google impl w/ ported weights

efficientdet. (Pretrained weights provided) EfficientDet: Scalable and Efficient Object Detection implementation by Signatrix GmbH

insightface. State-of-the-art 2D and 3D Face Analysis Project

fast-autoaugment-efficientnet-pytorch. A Pytorch implementation of Fast AutoAugment and EfficientNet

hyperopt. Distributed Asynchronous Hyperparameter Optimization in Python

imgclsmob. Sandbox for training deep learning networks

Yet-Another-EfficientDet-Pytorch. The pytorch re-implement of the official efficientdet with SOTA performance in real time and pretrained weights.

latex-action. :octocat: GitHub Action to compile LaTeX documents

pytorch-HED. An implementation of HED in pytorch

manifold_mixup. Code for reproducing Manifold Mixup results (ICML 2019)

mask-rcnn-tensorflow. Fork of Tensorpack to make breaking performance improvements to the Mask RCNN example. Training is approximately 2x faster than the original implementation on AWS.

tensorpack. A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility

RFConv. Rectified Convolution

detectron2-ResNeSt. A fork of Detectron2 with ResNeSt backbone

regnet. Pytorch implementation of network design paradigm described in the paper "Designing Network Design Spaces"

cutmix. a Ready-to-use PyTorch Extension of Unofficial CutMix Implementations with more improved performance.

CutMix-PyTorch. Official Pytorch implementation of CutMix regularizer