Wulingtian

yolov5_tensorrt_int8_tools

tensorrt int8 量化yolov5 onnx模型

yolov5_tensorrt_int8

TensorRT int8 量化部署 yolov5s 模型，实测3.3ms一帧！

yolov5_onnx2caffe

yolov5 onnx caffe

yolov5_caffe

yolov5 onnx caffe

RepVGG_TensorRT_int8

RepVGG TensorRT int8 量化，实测推理不到1ms一帧！

EfficientNetv2_TensorRT_int8

EfficientNetv2 TensorRT int8

nanodet_tensorrt_int8

nanodet int8 量化，实测推理2ms一帧！

nanodet_openvino

手把手教你OpenVINO下部署NanoDet模型，intel i7-7700HQ CPU实测6ms一帧

nanodet_tensorrt_int8_tools

nanodet

DiverseBranchBlock_TensorRT_int8

data-science-competition

该仓库用于记录和定期提供各大数据科学竞赛的赛事消息和原创baseline，思路分享以及博主的一些竞赛心得和学习资料等. 主要涵盖：kaggle, 阿里天池，华为云大赛校园赛，百度aistudio，和鲸社区，datafountain等

DDRNet

The official implementation of "Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes"

Deepsort_V2

2020中兴捧月阿尔法赛道多目标检测和跟踪初赛第一名方案

fast-reid

SOTA Re-identification Methods and Toolbox

HarDNet-MSEG

Inpaint-Anything

Inpaint anything using SAM + inpainting models.

kaggle-global-wheat-detection

9th Place Solution of Kaggle Global Wheat Detection

micronet

micronet, a model compression and deploy lib. compression: 1、quantization: quantization-aware-training(QAT), High-Bit(>2b)(DoReFa/Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference)、Low-Bit(≤2b)/Ternary and Binary(TWN/BNN/XNOR-Net); post-training-quantization(PTQ), 8-bit(tensorrt); 2、 pruning: normal、regular and group convolutional channel pruning; 3、 group convolution structure; 4、batch-normalization fuse for quantization. deploy: tensorrt, fp32/fp16/int8(ptq-calibration)、op-adapt(upsample)、dynamic_shape

model_search

ncnn-android-yolov5

The YOLOv5 object detection android example

Pedestrian-detection-paper-list

mark useful paper

quarkdet

QuarkDet lightweight object detection in PyTorch .Real-Time Object Detection on Mobile Devices.

RepVGG_openvino

i7-7700HQ实测10ms一帧，包括前处理和后处理！

sahi

A vision library for performing sliced inference on large images/small objects

tensorRTIntegrate

TensorRT ONNX Plugin、Inference、Compile

Weighted-Boxes-Fusion

Set of methods to ensemble boxes from different object detection models, including implementation of "Weighted boxes fusion (WBF)" method.

Wulingtian

yolo_quantization

Based of paper "Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference"

yolov5-rt-stack

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