pytorch, hand(object) detect ,yolo v5，手检测

Last update: Dec 20, 2022

Related tags

Deep Learning yolo-v5

Overview

YOLO V5

物体检测，包括手部检测。

项目介绍

手部检测

手部检测示例如下：

视频示例：

项目配置

作者开发环境：
Python 3.7
PyTorch >= 1.5.1

数据集

手部检测数据集

该项目数据集采用 TV-Hand 和 COCO-Hand (COCO-Hand-Big 部分) 进行制作。
TV-Hand 和 COCO-Hand数据集官网地址 http://vision.cs.stonybrook.edu/~supreeth/

感谢数据集贡献者。    
Paper：  
Contextual Attention for Hand Detection in the Wild. S. Narasimhaswamy, Z. Wei, Y. Wang, J. Zhang, and M. Hoai, IEEE International Conference on Computer Vision, ICCV 2019.

该项目制作的训练集的数据集下载地址(百度网盘 Password: 25y3 )

所有数据集的数据格式

size是全图分辨率， (x，y) 是目标物体中心对于全图的归一化坐标，w,h是目标物体边界框对于全图的归一化宽、高。

dw = 1./(size[0])  
dh = 1./(size[1])  
x = (box[0] + box[1])/2.0 - 1  
y = (box[2] + box[3])/2.0 - 1  
w = box[1] - box[0]  
h = box[3] - box[2]  
x = x*dw  
w = w*dw  
y = y*dh  
h = h*dh

为了更好了解标注数据格式，可以通过运行 show_yolo_anno.py 脚本进行制作数据集的格式。注意配置脚本里的path和path_voc_names，path为标注数据集的相关文件路径，path_voc_names为数据集配置文件。

制作自己的训练数据集

如下所示,每一行代表一个物体实例，第一列是标签，后面是归一化的中心坐标(x,y),和归一化的宽(w)和高(h)，且每一列信息空格间隔。归一化公式如上，同时可以通过show_yolo_anno.py进行参数适配后，可视化验证其正确性。

label     x                  y                   w                  h
0 0.6200393316313977 0.5939000244140625 0.17241466452130497 0.14608001708984375
0 0.38552491996544863 0.5855700073242187 0.14937006832733554 0.1258599853515625
0 0.32889763138738515 0.701989990234375 0.031338589085055775 0.0671400146484375
0 0.760577424617577 0.69422998046875 0.028556443261975064 0.0548599853515625
0 0.5107086662232406 0.6921500244140625 0.018792660530470802 0.04682000732421875
0 0.9295538153861138 0.67602001953125 0.03884511231750328 0.01844000244140625

预训练模型

从零开始预训练模型

预训练模型下载地址(百度网盘 Password: ad4l )

手部检测预训练模型

包括yolo_v5预训练模型图像输入尺寸640。
预训练模型下载地址(百度网盘 Password: x7d4 )

项目使用方法

数据集可视化

根目录下运行命令： show_yolo_anno.py (注意脚本内相关参数配置 )

模型训练

根目录下运行命令： python train.py (注意脚本内相关参数配置 )

模型推理

根目录下运行命令： python video.py (注意脚本内相关参数配置 )

pytorch, hand(object) detect ,yolo v5，手检测

Related tags

Overview

YOLO V5

项目介绍

手部检测

项目配置

数据集

手部检测数据集

所有数据集的数据格式

制作自己的训练数据集

预训练模型

从零开始预训练模型

手部检测预训练模型

项目使用方法

数据集可视化

模型训练

模型推理

Owner

Eric.Lee

This is the official implementation of 3D-CVF: Generating Joint Camera and LiDAR Features Using Cross-View Spatial Feature Fusion for 3D Object Detection, built on SECOND.

🥇Samsung AI Challenge 2021 1등 솔루션입니다🥇

Code for “ACE-HGNN: Adaptive Curvature ExplorationHyperbolic Graph Neural Network”

Code for our TKDE paper "Understanding WeChat User Preferences and “Wow” Diffusion"

Code and data for paper "Deep Photo Style Transfer"

Earth Vision Foundation

Chinese Mandarin tts text-to-speech 中文 (普通话) 语音合成 , by fastspeech 2 , implemented in pytorch, using waveglow as vocoder,

Decorator for PyMC3

Computer Vision is an elective course of MSAI, SCSE, NTU, Singapore

Using machine learning to predict undergrad college admissions.

[CVPR 2021] Exemplar-Based Open-Set Panoptic Segmentation Network (EOPSN)

Binary Passage Retriever (BPR) - an efficient passage retriever for open-domain question answering

code for our paper "Source Data-absent Unsupervised Domain Adaptation through Hypothesis Transfer and Labeling Transfer"

Utilities and information for the signals.numer.ai tournament

3rd Place Solution of the Traffic4Cast Core Challenge @ NeurIPS 2021

Generative Adversarial Networks(GANs)

An Implementation of Fully Convolutional Networks in Tensorflow.

UniLM AI - Large-scale Self-supervised Pre-training across Tasks, Languages, and Modalities

Code for ICCV 2021 paper "HuMoR: 3D Human Motion Model for Robust Pose Estimation"

Multi-objective gym environments for reinforcement learning.

pytorch, hand(object) detect ,yolo v5，手检测

Related tags

Overview

YOLO V5

项目介绍

手部检测

项目配置

数据集

手部检测数据集

所有数据集的数据格式

制作自己的训练数据集

预训练模型

从零开始预训练模型

手部检测预训练模型

项目使用方法

数据集可视化

模型训练

模型推理

Owner

Eric.Lee

This is the official implementation of 3D-CVF: Generating Joint Camera and LiDAR Features Using Cross-View Spatial Feature Fusion for 3D Object Detection, built on SECOND.

🥇Samsung AI Challenge 2021 1등 솔루션입니다🥇

Code for “ACE-HGNN: Adaptive Curvature ExplorationHyperbolic Graph Neural Network”

Code for our TKDE paper "Understanding WeChat User Preferences and “Wow” Diffusion"

Code and data for paper "Deep Photo Style Transfer"

Earth Vision Foundation

Chinese Mandarin tts text-to-speech 中文 (普通话) 语音 合成 , by fastspeech 2 , implemented in pytorch, using waveglow as vocoder,

Decorator for PyMC3

Computer Vision is an elective course of MSAI, SCSE, NTU, Singapore

Using machine learning to predict undergrad college admissions.

[CVPR 2021] Exemplar-Based Open-Set Panoptic Segmentation Network (EOPSN)

Binary Passage Retriever (BPR) - an efficient passage retriever for open-domain question answering

code for our paper "Source Data-absent Unsupervised Domain Adaptation through Hypothesis Transfer and Labeling Transfer"

Utilities and information for the signals.numer.ai tournament

3rd Place Solution of the Traffic4Cast Core Challenge @ NeurIPS 2021

Generative Adversarial Networks(GANs)

An Implementation of Fully Convolutional Networks in Tensorflow.

UniLM AI - Large-scale Self-supervised Pre-training across Tasks, Languages, and Modalities

Code for ICCV 2021 paper "HuMoR: 3D Human Motion Model for Robust Pose Estimation"

Multi-objective gym environments for reinforcement learning.

Chinese Mandarin tts text-to-speech 中文 (普通话) 语音合成 , by fastspeech 2 , implemented in pytorch, using waveglow as vocoder,