Initializing... drag & drop files here
Supports: MTS
This tool pulls a single frame out of an AVCHD camcorder clip and saves it as an AVIF image. The output is one still photo, never a moving file — AVIF can technically hold an animation, but this conversion does not produce one. So the real question is whether AVIF is the right destination for a frame grab at all, or whether JPG, PNG or WebP suits you better. The short answer: AVIF if the still is going on a modern website or into long-term storage where size matters, JPG if it has to open in absolutely anything, PNG if it is going into an editor for further work.
| Property | AVIF | JPG | PNG | WebP |
|---|---|---|---|---|
| Underlying compression | AV1 intra-frame | DCT (JPEG) | DEFLATE | VP8/VP8L |
| Lossy or lossless | Lossy here | Lossy | Lossless | Both |
| Typical size for the same visual quality | Smallest | Largest of the four | Very large | Between AVIF and JPG |
| Global browser support | 94.67% (caniuse) | Universal | Universal | 96.18% (caniuse) |
| Browsers that added it | Chrome 85, Firefox 93, Safari 16.4, Edge 121 | Every browser ever | Every browser ever | Chrome 32, Firefox 65, Safari 16, Edge 18 |
| Opens in old desktop photo viewers | Often not | Yes | Yes | Sometimes |
| Windows Photos support | Needs the AV1 Video Extension | Built in | Built in | Built in |
| Colour depth ceiling | 10 and 12-bit capable | 8-bit | Up to 16-bit | 8-bit |
| Royalty-free | Yes (Alliance for Open Media) | Yes | Yes | Yes |
| Best for | Web delivery, archiving stills | Sharing with anyone | Editing, screenshots | Web delivery with wider reach |
A note on that colour-depth row: AVIF can store 10 and 12-bit images, but an AVCHD camcorder records 8-bit H.264, so a frame taken from your clip is 8-bit no matter which format you export to. Choosing AVIF does not add depth or dynamic range that the source never had, and there is no HDR tone-mapping step in this conversion.
.mts files; you can queue several clips and take a frame from each.The "Time (seconds)" field opens at 0, and on camcorder footage second zero is very often the worst frame in the entire clip. Handycams begin writing the moment you press record, so the opening frames typically catch the lens still hunting for focus, the auto-exposure still settling, or a black lead-in before the sensor image arrives. Scrub the clip in any player first, find a moment where your subject is sharp and still, and type that number in.
| Frame Selection option | What it does | Good for |
|---|---|---|
| Specific Frame + Time (seconds) | Seeks to that second and exports one still | The normal case — a chosen moment |
| Multiple Screenshots + Capture Rate | Exports evenly spaced stills across the clip | Contact sheets, finding the best frame |
Motion matters as much as timing. Most AVCHD camcorders record 1080i, where each frame is woven from two fields captured a fraction of a second apart. On a moving subject those fields disagree and the extracted still shows horizontal comb lines. Nothing in this pipeline deinterlaces, so the only reliable fix is to choose a low-motion timestamp — a held pose, a tripod shot, the moment before someone swings — rather than a mid-action frame.
No. This conversion extracts still frames only. You get a single image from the timestamp you set, or a set of separate images if you switch to "Multiple Screenshots" — never a moving AVIF. If you want motion, convert the clip to a video format such as MP4 instead.
Because the timestamp is still at its default of 0 seconds. Camcorders start recording before the lens has finished focusing and before auto-exposure settles, so the first moments of an AVCHD clip are frequently soft, dark or blank. Enter a timestamp a few seconds into the clip and the problem usually disappears.
That is interlacing, not a conversion fault. AVCHD camcorders commonly shoot 1080i, building each frame from two fields taken moments apart, so anything that moved between them lands in two places. There is no deinterlacing step in this conversion, and exporting to a different image format will not remove it. Pick a low-motion timestamp, or deinterlace the video in an editor before grabbing frames.
In browsers, almost certainly — caniuse currently puts global AVIF support at 94.67%, with Chrome 85+, Firefox 93+, Safari 16.4+ and Edge 121+ all handling it. Desktop applications lag behind: Windows Photos needs the free AV1 Video Extension installed, and plenty of older photo viewers and editors still cannot open AVIF at all. If the file needs to work everywhere without fuss, take the frame as JPG instead.
Generally yes, and often substantially so, because AVIF uses AV1's intra-frame coding rather than the 1990s DCT approach JPEG uses. The exact ratio depends entirely on the image — fine detail like foliage or a crowd compresses differently from a smooth sky. The trade is size against compatibility, not size against quality.
No, and there is nothing to keep. AVCHD records 8-bit H.264, so the source frame is 8-bit. AVIF as a format supports 10 and 12-bit images, but this conversion cannot manufacture depth the camera never captured, and there is no tone-mapping step in the pipeline.
By default it matches the source frame, because Image resolution opens on "Keep original" — normally 1920×1080. You can switch to a preset, a percentage, or explicit dimensions. Be aware the presets set height only, with width following the source aspect ratio, so on a 1440×1080 anamorphic AVCHD clip a preset will not restore the intended 16:9 shape. Use Width x Height and enter 1920 x 1080 for that.
This page accepts .mts. Clips pulled off a camcorder by software are frequently renamed .m2ts, and although the stream inside is the same, the two extensions are handled by separate pages. Either rename the file back to .mts or use the M2TS equivalent.
Your .MTS file is uploaded over an encrypted connection and the frame is extracted on our servers. Uploaded files and their results are deleted automatically after a few hours, are never shared or made public, and there is no sign-up and no watermark.