We have been working on:
- Collecting images based on provided keywords
- Annotate images, with various types of shapes & lines, for various kinds of objects (cars, human, animals, trees, etc.)
- Label images, following various rules (occluded, intact, invisible, etc.)