FrameBolt
Turn a whole video into images and get them back as one ZIP. There is no cap on how many frames: the archive is written as it goes, so five minutes at 30 fps is 9,000 pictures and not a dead tab.
A whole video, turned into pictures
Give it a video and it walks through it frame by frame, encodes each one and writes it straight into a ZIP. Nothing is uploaded, nothing piles up in memory, and nothing puts a ceiling on how many frames come out.
Written as it goes, not piled up
Extractors that run in a browser usually hold every image in memory and build the ZIP at the end, which is exactly why they stop at a few hundred frames. Here each frame is encoded, appended to the archive and dropped: all that survives a frame is a 60-byte entry in the index.
What it is for, and what it never does
Datasets for training and annotation, sprite sheets, contact sheets for an edit, hunting for the one frame where the sign is readable, a thumbnail every ten seconds of a long recording, or checking an animation frame by frame.
Your browser decodes the video and your own machine builds the ZIP. There is no server in the middle, no account and no history: close the tab and nothing of it is left anywhere.
Frequently asked questions
QHow many frames can it really get out?+
As many as the video has. Each image is written into the ZIP and forgotten, so what is in memory at any moment is one frame, not all of them. Sending the archive straight to disk removes the last ceiling: from there on the limit is the free space on your drive.
QDoes the video get uploaded anywhere?+
No. The decoding, the encoding and the ZIP all happen inside your browser. Nothing travels: there is no server that could receive the file, and no account or history to record that you opened it.
QHow long does it take?+
Roughly the length of the video for 'every frame', because it walks it in order and stops on each frame to write it. Picking one in every N or one every so many seconds is much faster, because those modes jump to the instants you asked for instead of crossing the whole video.
QWhich format should I choose?+
JPEG for almost everything: it is a fraction of the size and the difference is invisible on video, which is compressed already. WebP is smaller still at the same quality. PNG only if you need it lossless, and be ready for an archive several times bigger.
QWhat is the difference with FrameSnap?+
FrameSnap is a viewer for choosing THE frame: you step through the video, compare and export the one you want. FrameBolt does not show you a gallery, and that is the point — nothing is held on screen, so it can pull out tens of thousands of frames in one go.