How to Extract a Frame From a Video as a Full-Quality Image
Save a frame from a video as a sharp JPG or PNG at full resolution. Why screenshots look worse, and how to step to the exact frame you want.
Updated October 14, 20266 min read
To extract a frame from a video, open the video in a frame grabber, move to the moment you want, step forward or back one frame at a time until the picture is sharp, and save it as PNG or JPG. The result is the frame itself at the video's own resolution, not a screenshot of your screen. That difference matters more than people expect: a screenshot of a paused 4K video on a phone can come out at a fraction of the resolution, with the player's controls in it. This guide explains how to get a clean still, how to pick the sharpest frame, and how to save several frames in one go.
Why a screenshot is the wrong tool
Pausing a video and taking a screenshot is the obvious method, and it has three problems:
- Resolution. A screenshot captures your screen, not the video. A 4K video (3840 × 2160) shown in a window 1280 pixels wide gives you a 1280-pixel image, and you have thrown away most of the detail.
- Overlays. Pause icons, progress bars, captions and the system clock end up in the picture.
- Imprecision. Pausing lands wherever your tap or click happened to land, often on a frame blurred by motion.
Extracting the frame decodes the actual picture stored in the file, at its full stored size, with nothing on top. Some apps have this built in — Google Photos on Android offers an Export frame option while editing a video, and the VLC player has a snapshot command — but they rarely let you step frame by frame or choose the output size and format.
Step by step
Add the video
Drop an MP4, MOV, WebM, MKV or animated GIF into the frame grabber. It is read on your device; nothing is uploaded.
Scrub to roughly the right moment
Drag the capture-time slider. The preview shows the finished image, with the same crop, size and format the file will have.
Step to the exact frame
Use the one-frame-back and one-frame-forward buttons. Each step is one frame at the video's own rate, so a 24 fps film and a 60 fps screen recording both move by exactly one picture.
Choose the size and format
Pick Original size to keep the video's own resolution, or a preset such as YouTube 1280 × 720. Choose PNG, JPEG or WebP.
Capture and download
Press Capture and save the image. Several captures download together as a single ZIP.
Finding the sharpest frame
Video is recorded for motion, not for stills. At normal shutter speeds, anything moving is slightly blurred in each frame, and the frames either side of the one you want often contain a half-closed eye or a hand mid-gesture. A few habits help:
- Look for the pause in the movement. The sharpest frames come at the end of a gesture, when a head turn stops or a ball reaches the top of its arc.
- Step several frames either side. Three or four frames earlier or later is a fraction of a second but can be the difference between blurred and sharp.
- Avoid the first and last second. Clips often fade up from black or down to it.
- Compare at thumbnail size. If the still is going to be a thumbnail, judge it small; a picture that works full-size may not read at 200 pixels wide.
If even the best frame is soft, the image enhancer can lift contrast and dull colour, and fixing blurry photos explains what sharpening can and cannot rescue.
Step to the exact frame and save it at full resolution as PNG, JPEG or WebP. It all runs on your device, not our servers.
Extract a frame from a videoPNG, JPG or WebP?
| Format | Best for | Trade-off |
|---|---|---|
| PNG | Screen recordings, text, slides, graphics | Larger file; no further quality loss |
| JPEG | Camera footage, people, scenery | Much smaller; slight compression |
| WebP | Web pages where size matters most | Usually the smallest; a few older apps refuse it |
The frame was already compressed once by the video encoder, so a PNG cannot add quality back — it simply avoids losing any more. For a frame from camera footage, a high-quality JPEG looks the same at a fraction of the size. For a screen recording with small text, PNG keeps the edges clean. The broader comparison is in PNG vs JPG vs WebP.
What resolution will the image be?
At most, the resolution of the video. A 1080p video gives a 1920 × 1080 still and a 4K video gives 3840 × 2160 — about 8.3 megapixels, which prints well at up to about 13 × 7 inches at 300 DPI. A 720p video gives only 1280 × 720, roughly a 0.9-megapixel photo, fine on a screen and too small for most prints.
If you ask for a size larger than the source, the grabber treats it as a ceiling rather than enlarging the frame, because enlarging cannot add detail that was never recorded, and it tells you when that has happened. You can override it with Allow enlarging past the source when a platform demands a minimum size. If you need a genuinely larger image, the image upscaler does a much better job than a plain resize. On the free plan the exported still is capped at 1920 pixels on its longest side; Pro exports at the source's full resolution, including 4K.
Grabbing many frames at once
Sometimes you want options rather than one frame: candidates for a thumbnail, stills for a storyboard, or a quick visual summary of a long recording. Ask for several captures and each gets its own time slot, or press Space evenly to spread them across the whole clip. All the requested moments are read in a single pass through the video, so several captures cost barely more time than one.
A contact sheet tiles every capture into one image, like the storyboard grid in a media catalogue. It is the fastest way to review a long recording at a glance, or to show someone what a clip contains without sending the clip.
The free plan takes one video at a time, up to 300 MB and an hour long, and captures up to five frames per run. Pro captures up to 50 frames per run, adds the contact sheet, and takes up to 10 videos in one batch.
Using the frame as a thumbnail
The most common reason to grab a frame is a video thumbnail. The grabber has presets for the sizes platforms ask for: YouTube 1280 × 720, Shorts and Reels covers at 1080 × 1920, square 1080 × 1080, and link previews at 1200 × 630. Choose Crop to fill the box and trim the overflow, or Letterbox to fit the whole frame and pad the rest. To add a headline or an arrow, take the frame into the YouTube thumbnail maker. The design rules that make thumbnails get clicked are in YouTube thumbnail size.
If what you actually want is the movement — a reaction, a short loop — save a few seconds as an animation instead: see converting video to GIF.
Frequently asked questions
How do I extract a frame from a video?
Open the video in the frame grabber, scrub to the moment you want, step one frame at a time until it is sharp, choose PNG or JPEG, and press Capture.
How do I get a high-quality still from a video?
Extract the frame rather than taking a screenshot. Extraction decodes the picture at the video's own resolution with no player controls on top. Choose Original size and PNG for the cleanest result.
Can I extract a frame from a video on iPhone?
Yes. The frame grabber runs in Safari on iPhone and opens videos from your camera roll, including MOV files. Choose Original size to keep the full resolution.
What resolution will the extracted image be?
The same as the video at most: 1920 × 1080 from a 1080p video, 3840 × 2160 from 4K. The free plan caps exports at 1920 pixels on the longest side; Pro keeps 4K.
Can I save every frame of a video?
The grabber is built for chosen moments: up to five per run on the free plan and 50 on Pro, spaced by hand or evenly. Saving thousands of frames from a long video creates a folder far larger than the video itself.
Is my video uploaded?
No. The video is read and decoded on your own device, and the images are written there too.