Scene detection demo
LeanScene finds where each shot (one unbroken take from a single camera) ends and the next begins: hard cuts, and gradual transitions such as dissolves and fades. Drop a video and see every one on a timeline, with the frame before and after.
Loading model…
New: search the shots by description, with a CLIP model that also runs in this tab (an 88 MB download).
Shot search is on. Drop a video; when it has been processed, a search box appears above the player and you can find shots by describing them.
Starting…
Transitions
Each card shows the last frame before the transition and the first frame after it. Click a card to play from the transition.
Higher F1 than TransNet V2 on all three of its benchmarks, with 115× fewer operations per frame, in a 1.6 MB model.
Accuracy
F1 on the three standard shot-boundary benchmarks, scored the same way for both models
Cost
Per frame, from zero. Shorter is better
Operations (multiply-adds)
Parameters
How it works
- Every frame is decoded and shrunk to 100×100 pixels.
- A small network turns each frame into a short description, once.
- A second network reads those descriptions in order and labels each frame same, cut or fade.
Both run on your graphics card through WebGPU, or on the CPU where that is missing.
Use it yourself
Python
pip install git+https://github.com/lucasjacks0n/LeanScene leanscene video.mp4
Node.js
npm install leanscene npx leanscene video.mp4
Code, weights and training labels on GitHub · npm. The paper is in preparation.
This page reads MP4 and MOV files in a current Chrome, Edge or Safari.