Framejury, a culling assistant that works offline and explains every photo it rejects
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle What I Built Framejury looks through a folder of event photos, groups the frames that were shot seconds apart and look nearly the same, and marks the best one of each group to keep, while every other frame is rejected with a sentence that says why, the numbers behind that sentence are one click away, and a…
Framejury is an open-source offline photo culling assistant that analyzes event photos to group similar shots, select the best frame for each group, and provide a reason for every rejection. The tool runs on the photographer's computer using Ollama, ensuring no client photos are uploaded anywhere. Framejury offers attention as its primary value, allowing photographers to process thousands of frames in a fraction of the time it would take a human to manually select the best shots.
The analysis takes between 27 and 36 seconds for seven photos on a laptop with an RTX 4050 GPU. The tool provides a transparent explanation for each decision, including sharpness relative to the best frame, clipped areas, and what the model observed. Users can modify the assumption of four seconds of manual work per photo to estimate the time saved.
Framejury is built using Python, Node.js, and Ollama, and it has a Next.js 16 interface with React 19 and Tailwind 4. The project includes 42 tests that can run without the model and was first developed on October 13.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
