The Triton backend that allows running GPU-accelerated data pre-processing pipelines implemented in DALI's python API.

Official implementation of Aurora

🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch and FLAX.

LLM in Triton , Hugging Face -> Pytorch -> ONNX -> TensorRT -> Triton

Stable Diffusion web UI

Xtreme1 is an all-in-one data labeling and annotation platform for multimodal data training and supports 3D LiDAR point cloud, image, and LLM.

RepVGG: Making VGG-style ConvNets Great Again

A project demonstrating Lidar related AI solutions, including three GPU accelerated Lidar/camera DL networks (PointPillars, CenterPoint, BEVFusion) and the related libs (cuPCL, 3D SparseConvolution…

Source code for Twitter's Recommendation Algorithm

[NeurIPS 2022] 4D Unsupervised Object Discovery

[CVPR'22] Lite Pose: Efficient Architecture Design for 2D Human Pose Estimation

(WIP)The deployment framework aims to provide a simple, lightweight, fast integrated, pipelined deployment framework for algorithm service that ensures reliability, high concurrency and scalability…

Codebase for Image Classification Research, written in PyTorch.

利用pytorch实现图像分类的一个完整的代码，训练，预测，TTA，模型融合，模型部署，cnn提取特征，svm或者随机森林等进行分类，模型蒸馏，一个完整的代码

VR-Caps: A Virtual Environment for Active Capsule Endoscopy

Learning embeddings for classification, retrieval and ranking.

(ICCV'19 Best Paper Nomination) Larger Norm More Transferable: An Adaptive Feature Norm Approach for Unsupervised Domain Adaptation

特征提取/数据降维:PCA、LDA、MDS、LLE、TSNE等降维算法的python实现

Official Implementation of the WACV 2022 paper: Improve Object Detection with Label Assignment Distillation.

PoseFlow: Efficient Online Pose Tracking (BMVC'18)

Avalanche: an End-to-End Library for Continual Learning based on PyTorch.

Localization Distillation for Object Detection (CVPR 2022, TPAMI 2023)

Unofficial pytorch implementation for Self-critical Sequence Training for Image Captioning. and others.

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

We present MocapNET, a real-time method that estimates the 3D human pose directly in the popular Bio Vision Hierarchy (BVH) format, given estimations of the 2D body joints originating from monocula…

yolort is a runtime stack for yolov5 on specialized accelerators such as tensorrt, libtorch, onnxruntime, tvm and ncnn.

More readable and flexible yolov5 with more backbone(gcn, resnet, shufflenet, moblienet, efficientnet, hrnet, swin-transformer, etc) and (cbam，dcn and so on), and tensorrt

[NeurIPS'21] Unified tracking framework with a single appearance model. It supports Single Object Tracking (SOT), Video Object Segmentation (VOS), Multi-Object Tracking (MOT), Multi-Object Tracking…

Mengzi Pretrained Models