danielrosehill/accidental-and-low-quality-photos

Accidental and Low-Quality Photos Dataset

This repository contains unintentional photos from camera rolls - accidental captures, blurry shots, and other non-intentionally captured images.

Purpose

The intended use case is training an image organization model to automatically distinguish and suggest for deletion non-intentionally captured images. This can help users efficiently clean up their camera rolls by identifying photos that were taken accidentally or are of poor quality.

Dataset

This collection represents the start of a small image repository containing examples of:

• Accidental captures (pocket shots, finger over lens, etc.)
• Blurry or out-of-focus images
• Unintended duplicates or burst mode mistakes
• Other low-quality images that users typically want to remove

Use Cases

• Training classification models to detect accidental photos
• Developing photo management tools with auto-cleanup suggestions
• Research on image quality assessment
• Building automated camera roll organization systems