back

cineflow

an editor spends most of their time deciding where to cut. cineflow reads the footage and proposes those points, then hands them back fast enough to feel like part of the timeline. 1st place at the adobe express hackathon.

A strip of video frames with three proposed cut points marked between shots, scored by the model. a clip, frame by frame cut cut cut the model scores every boundary between frames and keeps the peaks
the output is a list of timestamps, not a rendered video. the editor keeps the last word.

the problem

cut detection models are usually trained to find hard scene changes, which is a different question from where a cut belongs. a hard change is a fact about pixels. a good cut is a fact about motion, framing, and pacing, and it often sits a few frames off the obvious one.

Relative F1 on cut detection: the C3D baseline at 1.00 and the fine-tuned vision transformer 23 percent higher at 1.23. f1 on held-out clips, relative to the baseline c3d baseline 1.00 cineflow 1.23 0
23% higher f1 than a c3d baseline on the same held-out clips.

how it works

what it means

the accuracy number matters less than the latency one. a suggestion that arrives after the editor has already made the cut is worth nothing, so the whole design is bent around answering before they finish moving the playhead.