Git Product home page Git Product logo

yolov8-pose's Introduction

YOLOv8-pose re-implementation using PyTorch

Installation

conda create -n YOLO python=3.8
conda activate YOLO
conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch-lts
pip install opencv-python==4.5.5.64
pip install PyYAML
pip install tqdm

Train

  • Configure your pose dataset path in main.py for training
  • Run bash main.sh $ --train for training, $ is number of GPUs

Test

  • Configure your dataset path in main.py for testing
  • Run python main.py --test for testing

Results

Version Epochs Pose mAP Download
v8_n_pose 1000 50.2 model
v8_n_pose* 1000 50.5 model
v8_s_pose* 1000 59.5 model
v8_m_pose* 1000 63.8 model
v8_l_pose* 1000 67.4 model
v8_x_pose* 1000 69.4 model
  • * means that weights are ported from original repo, see reference

Dataset structure

├── COCO 
    ├── images
        ├── train2017
            ├── 1111.jpg
            ├── 2222.jpg
        ├── val2017
            ├── 1111.jpg
            ├── 2222.jpg
    ├── labels
        ├── train2017
            ├── 1111.txt
            ├── 2222.txt
        ├── val2017
            ├── 1111.txt
            ├── 2222.txt

Results

Alt Text

Reference

yolov8-pose's People

Contributors

jahongir7174 avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar

yolov8-pose's Issues

Some problems in Dataset structure

Thank you for your work,
can you give more details about dataset structure, it looks different from the code which has train2017.txt and val2017.txt. I found possible txt files and put them under COCOPose, at the same level as images and labels, the content of train2017.txt is ./images/train2017/000000391895.jpg ...... (one per line). The content of val2017.txt is ./images/val2017/000000182611.jpg ...... (one per line).
In addition, there was an error. I modified line 583 of utils.py based on the yolov8 source code, changing traget_bboxes, target_scores, fg_mask,traget_gt_idx = assigned_targets to _, traget_bboxes, target_scores, fg_mask,traget_gt_idx = assigned_targets. I successfully ran the code, but BoxAP And PoseAP is still 0 after dozens of rounds of training, and it is also 0 after running the test code using the weights you provided. I would like to know whether it is a problem with the dataset or something else, and I look forward to your reply.

problem with dataloader

Hi, I got this error while running main.py, could you help me to solve it:

utils/dataset.py", line 121, in load_image
h, w = image.shape[:2]
AttributeError: 'NoneType' object has no attribute 'shape'

KPT_SIGMA

Hi, I want to add extra keypoint and wants to know how can I set value related to this key point in KPT_SIGMA array?

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.