How to extract frames from a video
- Drop a video file into the box.
- Play or scrub to the moment you want, then fine-tune with the frame buttons, the arrow keys or the ±1 second buttons.
- Click Capture PNG or Capture JPG to save that frame to the gallery below.
- For many frames, set Every N seconds and click Capture all, then download single frames or all of them as a ZIP.
What you can do with Video Frame Grabber
- Frame-by-frame stepping
- Step backward and forward by 1/24, 1/25, 1/30 or 1/60 of a second with buttons or the left and right arrow keys.
- Full-resolution stills
- Frames are captured at the video's native size, so a 4K video gives you 3840 × 2160 images, not a screen-sized copy.
- PNG or JPG
- PNG for lossless frames you plan to edit, JPG for smaller files. The video's native size is shown under the buttons.
- Interval capture
- Grab a frame every N seconds, as often as every 0.1 seconds, up to 300 frames per run, with a progress bar and cancel.
- Timestamped files
- Each frame is named with its exact time in seconds, and the ZIP numbers them in order for easy sorting.
Frame rates and getting sharp stills from video
A video is a fast sequence of still images, usually 24, 25, 30 or 60 per second. Pausing a video player and taking a screenshot captures only what fits on your screen, often scaled down and with player controls on top. Extracting the frame from the decoded video instead gives you the full native resolution with nothing overlaid. That is useful for picking a YouTube thumbnail from your own footage, pulling a sharp still from a phone video, documenting a moment for a bug report, or getting reference images for drawing and animation.
Match the step size to the video's frame rate for true frame-by-frame movement: 1/30 s for most phone video, 1/24 s for film-style footage, 1/25 s for European broadcast and 1/60 s for high-frame-rate clips. Motion blur is part of each frame, so for the sharpest still, look for a moment where the subject is briefly still. Interval capture is handy for time-lapses, storyboards and dataset building; a 10 minute video at one frame every 5 seconds gives 121 images. For a single overview image, the Video Info tool can build a contact sheet.
Frequently asked questions
Are my videos uploaded to extract frames?
No. The video is decoded by your browser's built-in player, frames are copied to a canvas, and PNG or JPG files are created on your device. Nothing is sent to a server, so it is safe for private or unreleased footage.
What resolution are the extracted frames?
The video's native resolution, shown as Native size after loading. A 1080p video gives 1920 × 1080 frames and a 4K video gives 3840 × 2160, regardless of how large the player looks on your screen.
How do I extract every frame of a video?
Interval capture goes down to one frame every 0.1 seconds (10 frames per second) and up to 300 frames per run, so it can't grab every frame of 24 to 60 fps video automatically. For every single frame of a short moment, step with the arrow keys and capture each one. For longer videos, use a larger interval.
Should I save frames as PNG or JPG?
PNG is lossless and best if you will edit the image, add text or need maximum sharpness, but files are large. JPG at the built-in high quality is a fraction of the size and fine for thumbnails, sharing and references. Interval captures default to JPG to keep ZIPs manageable.
Can I use it to make a YouTube thumbnail?
Yes. Find the best moment in your own video, capture it as PNG, then use the Image Resizer with the YouTube thumbnail preset (1280 × 720) and preview it with the YouTube Thumbnail Preview tool before uploading.