11 — Computer Vision¶
Computer vision: image processing, detection, segmentation, and real-time inference.
Scope¶
Cheat sheets for image and video pipelines — OpenCV, Pillow, torchvision, and YOLO/Ultralytics. Deep learning backbones (ResNet training, ViT) belong to 05-Deep-Learning.
Cheat Sheets¶
| Sheet | Covers |
|---|---|
| opencv.md | Image I/O, resize, color spaces, filtering, contours, video, webcam |
| pillow.md | PIL/Pillow: I/O, transforms, filters, drawing, batch processing |
| torchvision | Transforms, pretrained models, ImageNet inference, fine-tuning |
| yolo | YOLO detection concepts, training, inference, export |
| ultralytics | Ultralytics API: detection, segmentation, pose, tracking, CLI |
Contribute¶
- Template: CHEATSHEET_TEMPLATE.md
- Guide: CONTRIBUTING.md
- Planned: albumentations augmentation, instance/semantic segmentation deep-dive