- Document the computer vision projects that I have attempted so far
- Easy to use Python implementation of "Cut and paste" augmentation for object detection
- The main idea was taken from Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation
-
- This repo added the capability of saving images and annotations in XML annotation (Pascal VOC format)
- Reading XML Annotations and visualising it on image(s)
- Creating XML Annotations utilising PASCAL VOC Writer
- Resizing an image and its bounding box
- Document my journey in preprocessing publicly available dataset (ImageNet data) so that it fulfill the desired format in training our model (bags +persons)