视频全帧提取工具
从 MP4 或 WebM 视频中提取每一帧并以 PNG 或 JPEG 打包成 ZIP 下载。100% 在浏览器运行，无需上传，无大小限制。
100% client-side · your video never leaves your device
Extraction mode
to
Leave blank for full video Output format
85%
Load a video file to begin.
How to extract all frames from a video
- Click Choose video file and select your MP4, WebM, or MOV. The video appears in the preview so you can confirm the right file was loaded.
- Choose your extraction mode. Every N seconds samples the video at a fixed time interval — great for surveillance footage or timelapse review. Total N frames spreads a set number of frames evenly across the video — ideal for building training datasets. Optionally narrow the range with start/end fields, and pick PNG (lossless) or JPEG (smaller) as the output format.
-
Click Extract Frames. A progress bar updates per frame. When all frames are
captured, click Download ZIP to save the archive. Each frame is named with
its sequence number and timestamp (e.g.
frame-0001-1.00s.jpg).
Common use cases
- Machine learning datasets: Extract frames from screen recordings or real-world video to build image classification or object detection datasets — all without uploading to a third-party service.
- Video storyboards: Sample one frame per second from a presentation recording to create a visual script or slide deck summary.
- GIF creation: Extract a dense sequence of frames from a short clip, then import them into a GIF editor like the GIF Maker.
- Thumbnail selection: Review every frame to pick the sharpest, most expressive one to use as a video thumbnail or social media cover image.
Related tools
- Video Thumbnail Generator — extract a single frame at a specific timestamp via ffmpeg.wasm
- Video Trimmer — cut a clip before extracting frames
- GIF Maker — turn extracted frames into an animated GIF
- Millimeters to Points — typography unit converter
- Extract frames from MP4 videos
- Extract frames from WebM videos
常见问题
- Does my video leave my device?
- No. All frame extraction runs locally in your browser using the HTML Video element and Canvas API. Your video is never uploaded to any server.
- What video formats are supported?
- Any format your browser can play natively: MP4 (H.264), WebM (VP8/VP9/AV1), and MOV on most browsers. If the browser cannot decode the file, a clear error message is shown. Formats like AVI and MKV often require re-encoding to MP4 first.
- How many frames can I extract?
- Up to 300 frames per run to keep your browser responsive. Use "Total N frames" mode to pick exactly how many you need. For very long videos, extract at a wider interval to stay within the 300-frame limit.
- What image format should I choose — PNG or JPEG?
- Choose PNG when you need pixel-perfect lossless frames (e.g. for machine learning datasets or frame-accurate analysis). Choose JPEG for smaller files when visual quality is sufficient (e.g. storyboards or thumbnail selection). JPEG quality 85% is a good default.
- Why is frame extraction slow for some videos?
- Seeking in video requires the browser to decode back to the nearest keyframe, which can be slow for highly compressed or long-GOP files. Extracting at wider intervals (e.g. every 2–5 seconds) is significantly faster than dense per-second sampling.
- Can I extract frames from a specific time range?
- Set the start and end time in the controls to limit extraction to a section of the video. Use "Every N seconds" mode within that range for dense sampling, or "Total N frames" for evenly-spaced samples across the chosen section.
Last updated: By jarvisbox