Skip to content

edisonslightbulbs/PTOD-model-training

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Transfer learning using PyTorch's object detection API

Platform Hardware Dependencies
🔳 Linux 🔳 Azure Kinect 🔳 gflags
🔳 glog
🔳 Azure Kinect SDK
🔳 opencv
🔳 Anaconda
🔳 Yolov5
🔳 Image annotation tool

This project is made up of two sub-projects: image-capturing model-training. image-capturing is a CMake project that uses Microsoft's Azure Kinect to capture so-called depth color images (of cause, this can be changed). model-training uses shell and python scripts to exploit Tensor Flow's object detection API and train an object detection.


The notebooks in model-training README.md are self-documenting, but more on that in the model-training README.md. In principle, one can use any other camera, a webcam, or even already captured images (i.e., given a reasonable number of image captures exist) to train the detection model.