Convert
Frame / Thumbnail Extraction
Grab a single frame from a video at any timestamp and save it as a full-resolution PNG — right in your browser. Accepts MP4, MOV, WebM, MKV, and AVI. Nothing is uploaded. Free, no signup.
Drop a video file here, or click to browse
Everything runs in your browser — nothing is uploaded
What this is and why it matters
Sometimes the thing you need from a video is not a clip at all — it is a single picture. You want a thumbnail to represent a video before someone plays it, a poster image for a blog post, or a clean still of one exact moment: a child’s expression, a celebration, a scoreboard at the final whistle, a number plate, or a document someone held up during a call. A still image is shareable, embeddable, and printable in ways a video is not, and it is often the real deliverable.
The obvious alternative — taking a screenshot while the video plays — is fiddly and lossy. You have to pause at exactly the right instant, you capture only what fits on screen at whatever the player’s size is, and you usually pick up the player’s controls or browser chrome along the way. Extracting the frame properly instead gives you the image at the video’s full native resolution, framed exactly, with nothing extra around it, at precisely the timestamp you asked for.
This tool grabs the frame in the browser rather than on a server, for the same privacy reason as the rest of the toolkit. The video you want a still from is often personal — a family recording, a private call, an unreleased clip — and uploading the whole thing to a website just to get one picture back is slow and a privacy risk. Here the video is read into memory on your own device, the chosen frame is captured locally, and only that single PNG is produced. Nothing is transmitted and nothing is stored.
How to use this tool
Add your video file. Drag it onto the drop zone above, or click to open the file picker and choose a file from your device. MP4, MOV, WebM, MKV, and AVI are all accepted. The moment it is selected, the tool reads it locally and shows the file name, its size, and its total duration, so you know the range of times you can pick from. Nothing uploads — everything happens on your own machine. Files must be under 500 MB.
Enter the timestamp. Type the moment you want to capture. You can use plain seconds (for example 15), minutes and seconds (00:15), or hours, minutes and seconds (00:00:15), and you can include decimals for precision. Leave it blank to grab the very first frame. The tool checks that the time is within the video’s duration and tells you clearly if it is not.
Let the engine load on first use. The first time you capture a frame, the tool downloads a one-time engine of about 30 MB into your browser. This happens only once — afterwards it is cached and later captures start immediately.
Capture and download. Press Capture Frame and the tool grabs the frame at your timestamp and shows a preview of the image. If it is the one you want, download it as a full-resolution PNG named after your original video. If not, adjust the time and capture again — it takes only a moment.
Examples and use cases
Making a thumbnail for a video upload
A Bengaluru creator wants a thumbnail for a video they are uploading and would rather use a real frame than a separate photo. They type the timestamp of the best-looking moment, capture it as a full-resolution PNG, and use it as the poster image — all without uploading the video to any thumbnail website. Because the frame comes out at the video’s native resolution, it stays crisp when the platform scales it up for a large thumbnail, rather than looking soft the way a small screenshot would.
Saving a document shown on a video call
A Pune professional recorded a video call in which a colleague briefly held a printed figure up to the camera. They find the timestamp where it is clearest, capture that single frame, and save a readable still of the document — far cleaner than a hurried screenshot, and the recording never leaves their laptop.
Capturing a moment as a photo
A Chennai parent has a video of their child’s first steps and wants one frame as a printable photo. They enter the exact moment, capture it at the video’s native resolution, and get a clean PNG to print or share — keeping a precious private memory entirely on their own device.
Pulling one step from a recipe video
A Delhi home cook follows a long recipe video and wants a still of one particular step to keep alongside their notes. They type the timestamp of that step and save the frame as an image, building a small set of reference pictures without screenshotting or uploading anything. Repeating this for a few key steps gives them a tidy sequence of stills they can glance at while cooking, instead of scrubbing back and forth through the video with messy hands.
Frequently asked questions
- Is my video uploaded to a server?
- No. The entire capture happens inside your browser. Your video is read into memory on your own device, the chosen frame is extracted locally by a WebAssembly build of FFmpeg, and the PNG is handed straight back to you as a download — no byte of the video is ever sent to us or anyone else. This is especially valuable for personal recordings and private calls; the original never leaves your machine, and only the single image is produced. Close the tab and nothing is retained.
- What format is the image, and what resolution?
- The captured frame is saved as a PNG, which is lossless — it is an exact copy of the frame with no compression artefacts. The image comes out at the video’s own native resolution, so a frame from a 1080p video is a 1920 by 1080 PNG. That makes it suitable for thumbnails, poster images, printing, or anywhere you would use a high-quality still, without the loss or cropping you get from a screenshot taken while the video plays.
- How exact is the captured time?
- The tool seeks quickly to your timestamp for an instant capture, landing on the frame at or just before the time you entered. For grabbing a thumbnail or a still of a moment this is exactly what you want — it is near-instant and visually on the mark. If the very first frame you get is a hair off from the precise instant you wanted, simply adjust the time by a fraction of a second and capture again; each capture only takes a moment.
- Why does the first capture download about 30 MB?
- The capture is done by a full build of FFmpeg compiled to WebAssembly so it can run inside your browser. That engine is about 30 MB, and it downloads once the first time you use the tool. After that your browser caches it, so later captures start right away without downloading it again. This one-time cost is what lets the frame be extracted on your own device instead of on a server you would have to upload the video to.
- Is there a limit on file size?
- Yes — 500 MB for the input video. Because the work is done entirely by your own device inside a browser tab, the video has to fit in the tab’s memory while it is processed, and very large files would risk crashing the tab. For most clips this is comfortably enough; if your video is larger, trimming it down first or using a shorter source will bring it under the limit while still letting you grab the frame you want.