Guides

Practical guides on image formats, file sizes, photo metadata and computer vision datasets, with the browser tools to put them into practice.

• How to make a photo smaller (under 100 KB, 200 KB or 1 MB) without ruining it: Shrink a photo below 100 KB, 200 KB or 1 MB without visible damage: reduce the pixels first, then the quality, and change the format only if needed.
• JPG, PNG or WebP: choosing the right image format: JPG for photos that must open everywhere, PNG for exact graphics and transparency, WebP for smaller web images. The reasons, plus a decision table.
• What your photos reveal: EXIF, GPS and the other metadata in image files: Photos can carry GPS coordinates, serial numbers, edit history and even a hidden video. What EXIF stores, who strips it, and how to check and remove it.
• YOLO vs. COCO annotation format: fields, conversion and pitfalls: YOLO stores normalized center boxes in one txt file per image, COCO pixel boxes from the top-left corner in one JSON. Fields, conversion and pitfalls.
• From segmentation masks to YOLO and COCO polygons: How masks become YOLO and COCO polygons: connected regions, 4 vs. 8 neighbors, outline simplification, holes, touching objects, specks and file names.