Initializing... drag & drop files here
Supports: MTS
MTS is the filename an AVCHD camcorder writes straight to its card — Sony, Panasonic, Canon and JVC handycams all produce it, and the official AVCHD specification puts H.264/AVC video and Dolby Digital (AC-3) or Linear PCM audio inside an MPEG-2 transport stream. A .jfif file is an ordinary JPEG: JFIF is the interchange wrapper standardised as ITU-T T.871, approved in May 2011. This tutorial is for the person who wants one clean frame — a print, a thumbnail, a photo of a moment nobody took a photo of — without installing an editor.
The part most guides skip is which frame you get. Left alone, this tool captures the frame at second zero, and on camcorder footage second zero is very often a black lead-in, an auto-focus hunt, or the operator's hand still on the record button. Setting a timestamp is the single most important thing you do on this page.
.mts clip onto the page or pick it with the file browser. Clips copied straight off the camera's card work fine, and you can queue several at once.0. Type the moment you actually want — decimals such as 12.5 are accepted — or switch to Multiple Screenshots and choose a Capture Rate to sample the clip on a fixed cadence instead..jfif. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically a few hours later — no sign-up, no watermark, never shared or made public.Internally this is a two-stage job, and knowing that explains most of the behaviour. First the video stage seeks to your timestamp and writes exactly one frame out as a lossless PNG. Then the image stage encodes that PNG to JPEG and labels it .jfif. Nothing else touches the pixels in between, which is why the still carries the camera's original H.264 compression plus exactly one generation of JPEG — not two.
47.2 lands on the frame at 47.2 seconds rather than at a keyframe boundary somewhere nearby.0.1s seconds (single frame at 10fps) up to 10 seconds per frame; one frame per second over a 60-second clip gives you 60 candidate stills to pick from, each returned as its own JFIF.1920, or you will ask for a still nineteen times the original width.| Quality Preset | JPEG quality | Best for |
|---|---|---|
| Highest | 100 | Archival frame, print, or further retouching |
| Very High (opens selected) | 95 | The general-purpose choice; loss is hard to see at 100% zoom |
| High | 90 | Web-sized stills where the file still has to look clean |
| Medium | 80 | Thumbnails and contact sheets |
| Low | 70 | Preview grids where size beats fidelity |
| Very Low | 60 | Rough proofs only |
| Lowest | 50 | Smallest file; visible blocking on skies, walls, and gradients |
0. Camcorders routinely write a dark or out-of-focus lead-in at the head of a clip. Set Time (seconds) a few seconds in and convert again.1920 x 1080 in Pixels to get the intended proportions back..mts extension only. The byte-identical .m2ts spelling that Windows and playback software often produce has its own page — use M2TS to JFIF for those, or rename the file..jpg. The bytes are unchanged and every image viewer will open it — see the first question below for why.| Property | MTS (AVCHD source) | JFIF (output) |
|---|---|---|
| Standard | AVCHD, co-promoted by Sony and Panasonic | ITU-T T.871, approved May 2011 |
| Payload | H.264/AVC video plus AC-3 or Linear PCM audio | JPEG-compressed still image data |
| Container | MPEG-2 transport stream | JPEG stream with a JFIF APP0 marker segment |
| Bitrate ceiling | 24 Mbps, or 28 Mbps for stereoscopic recordings | Not applicable |
| Typical frame size | 1920x1080, or 1440x1080 in anamorphic modes | The source frame size unless you resize |
| Scanning | Progressive or interlaced, depending on the mode | A single instant; interlace combing is preserved as-is |
| Audio | Yes | None — the soundtrack is discarded |
This page produces images and nothing else. If you want the clip's soundtrack, MTS to M4A extracts it; if you want a playable copy of the whole video, MTS to MP4 is the sensible target. A still also can't undo what the camera did — a frame shot at 1/30s during a fast pan is motion-blurred in the source and no setting here will sharpen it. And if a clip was copied off the card while the camera was still writing, the tail can be truncated; re-copy the original file from the camcorder rather than trying to salvage the partial one.
Yes. JFIF is the JPEG File Interchange Format — the small APP0 header that sits just after a JPEG's start-of-image marker and declares things like pixel density. It was standardised as ITU-T T.871 in May 2011, long after it had become the de-facto way JPEGs are written. The compressed image data in a .jfif is ordinary JPEG data, so renaming the file to .jpg changes nothing at all and is completely safe. If you would rather not do the renaming, MTS to JPG produces the same picture under the more familiar extension.
Because Time (seconds) opens at 0 and the first frame of a camcorder clip is frequently unusable. AVCHD recordings often begin with a dark frame while exposure settles, an auto-focus hunt, or a moment before the subject is actually in shot. Set a timestamp a few seconds into the clip. If you are not sure where the good frame is, run Multiple Screenshots at one frame per second first, then re-run Specific Frame at the timestamp you liked.
No, and it is important to be straight about that: this pipeline does not deinterlace at any stage. An interlaced frame stores two fields captured a fraction of a second apart, so a moving edge shows the characteristic horizontal teeth, and that pattern is written into the JFIF exactly as it exists in the source. There is no preset, resolution, or quality setting on this page that removes it — nor does routing the clip through an MP4 conversion first. Your options are to choose a frame with little motion, or to shoot progressive in the first place.
No — it picks up exactly one JPEG generation here. The video stage decodes your chosen frame and writes it as a lossless PNG, and only then does the image stage encode it to JPEG. So the still carries the camcorder's original H.264 compression plus one round of JPEG at whatever quality your preset selects. If you want no JPEG generation at all on top of the camera's, use MTS to PNG instead and set its Quality Preset to Highest, which is the true-colour lossless setting on that page — the lower presets there apply palette quantization rather than storing every colour.
That mode is anamorphic: it stores 1440 non-square pixels per line that players stretch to a 1920-wide 16:9 picture at display time. When the still is resized here the pixel aspect is normalised to square, so the stored 1440x1080 shape is what you get and the image looks narrow. Fix it by choosing Width x Height under Image resolution, switching the unit dropdowns from Percent (%) to Pixels, and entering 1920 by 1080.
Yes — that is what Multiple Screenshots is for. Choose it under Frame Selection and set a Capture Rate; the list runs from ten frames a second down to one frame every ten seconds. Every captured frame is encoded as its own JFIF at whatever Quality Preset and resolution you selected, so a 90-second clip sampled once per second returns 90 images. Specific Frame is the mode that returns exactly one.
Because the accepted-extension list here is .mts only. It is an annoying distinction, because the streams are usually byte-identical — cameras write .mts on the card and Windows, Blu-ray tooling, and several import utilities write the same data as .m2ts. Either rename the file to .mts, or use the dedicated M2TS to JFIF page, which accepts the other spelling.
1920x1080 — a shade over two megapixels — because AVCHD is an HD format whose recording modes top out there, and a still cannot contain detail the video never captured. Keep original gives you that full frame. The Preset Resolution list goes up to 4320p (8K), but selecting anything above the source only interpolates pixels: the file gets bigger and the picture does not get sharper.
The practical limit is upload size and time rather than a hard cap on our side — AVCHD is a high-bitrate format, so a long clip is a big file and a slow connection is what usually hurts. It is worth trimming first if you only need one frame from deep inside a long recording: Video Cutter will cut the section out before you convert. Whatever you upload travels over an encrypted connection, is processed on our servers, and is deleted automatically a few hours after the job finishes. Nothing is shared or made public, and there is no sign-up or watermark.