Skip to content

[YOLOR + Mediapipe] Let the machine sees what object we are pointing to - RoboCup 2022

Notifications You must be signed in to change notification settings

robocup-eic/robocup2022-cv-what-is-that

 
 

Repository files navigation

robocup2022-cv-what-is-that

Computer Vision models for "What Is That" task in RoboCup@Home 2022
ref : https://www.augmentedstartups.com/YOLOR-You-Only-Learn-One-Representation-PyTorch

Installation

  1. create conda environment using python 3.9.6
  2. Install pytorch
    $ pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio===0.9.0 -f https://download.pytorch.org/whl/torch_stable.html
  3. Install Cython
    $ pip install -U cython
  4. Install pycocotool
    $ pip install "git+https://github.com/philferriere/cocoapi.git#egg=pycocotools&subdirectory=PythonAPI"
  5. Download yolor config folder from https://drive.google.com/file/d/17JveNhnGDXN2ilw_lLpfy6XLt7nj-sjd/view
  6. Put all downloaded files to /object_detection_module/config
  7. Install requirement.txt
    pip install -r requirements.txt

About

[YOLOR + Mediapipe] Let the machine sees what object we are pointing to - RoboCup 2022

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%