M2TS to HEIF Converter

M2TS to HEIF: One Frame Out of an AVCHD or Blu-ray Stream

.m2ts is a BDAV transport stream — the packet format Blu-ray discs use and AVCHD camcorders write to memory cards. HEIF is a still-image format, so this page does not convert the recording; it decodes one frame and saves that picture. Two things about camcorder footage in particular make this less routine than it sounds: a lot of AVCHD is interlaced, and some of it is anamorphic, which produces a squashed still unless you correct it. Both are covered below, along with the honest question of whether HEIF is the format you want at all.

How to Convert M2TS to HEIF

1. Upload Your M2TS File: Drag the .m2ts onto the page or click "+ Add Files". This page accepts the .m2ts spelling only — camcorder files still named .MTS are the identical stream, so either rename them or use MTS to HEIF. There is no fixed size limit; on a long HD recording the upload is the slow part, not the frame grab.
2. Set Frame Selection and Time (seconds): Frame Selection opens on Specific Frame with Time (seconds) at 0. Type the moment you want — decimals are accepted, so 47.5 is forty-seven and a half seconds in. Multiple Screenshots with a Capture Rate samples the whole clip instead, one file per captured frame.
3. Set Image resolution and Quality Preset: Image resolution opens on Keep original — but see the anamorphic section below before you leave it there. Quality Preset opens on Very High (Recommended), and Specific file size is available if you have a byte ceiling rather than a quality target.
4. Convert and Download: Click Convert and save the .heif. Files upload over an encrypted connection, are processed on our servers, and are deleted automatically after a few hours — no sign-up, no watermark, never shared or made public.

What Is Inside an .m2ts

The Anamorphic Trap: 1440 x 1080 Footage Comes Out Squashed

Not all AVCHD is stored with square pixels. A common recording mode stores a 16:9 picture in a 1440 x 1080 frame and marks it with a 4:3 pixel aspect ratio, so a player stretches it back out to 1920 x 1080 on screen. Still images have no equivalent field: PNG and HEIF both assume square pixels, and there is nowhere to record "these pixels are wider than they are tall".

We built a 1440 x 1080 clip flagged with a 4:3 sample aspect ratio, confirmed a player reads it as 16:9, then pulled a frame the way this page does. The still came back 1440 x 1080 — geometrically correct pixel data, but a quarter narrower than the shot looks in a player. Circles come out as ovals and faces look thin.

The fix is on this page. Switch Image resolution to Width x Height and type 1920 and 1080; the frame is stretched back to the shape the footage was meant to have. Two notes on doing that well:

• Check first, do not assume. Most 1920 x 1080 AVCHD is square-pixel and needs nothing. Open the clip in a player and compare the reported resolution with how it looks — if the player says 1440 x 1080 but the picture fills a 16:9 window, you have anamorphic footage.
• Do it in the conversion, not afterwards. Stretching a 1440-wide still to 1920 later re-samples an already-encoded image; doing it here means the resize happens on the lossless frame before the HEIF encoder ever runs.

HEIF, JPG or PNG for a Camcorder Frame

Our HEIF output is 8-bit with 4:2:0 chroma, which we confirmed on the encoded file. That is not a limitation you will feel on AVCHD footage, because consumer H.264 camcorder recordings are 8-bit 4:2:0 already — the still inherits exactly the colour resolution the video had. It does mean HEIF has no headroom advantage over M2TS to JPG here; the advantage is purely in file size.

Frequently Asked Questions

Why does my still have comb lines through moving objects?

Because the footage is interlaced and nothing in this pipeline de-interlaces. A great deal of AVCHD is recorded as 1080i, which stores two half-height fields captured a fraction of a second apart and woven into one frame. On a static shot you will never notice; on anything moving, the two fields disagree and you get the horizontal comb pattern. There is no setting here that removes it — pick a frame from a still moment, or de-interlace the clip in a video editor before extracting.

My 1440 x 1080 camcorder frame looks squashed. What happened?

Your footage is anamorphic. It stores a 16:9 picture inside a 1440-pixel-wide frame and relies on a pixel-aspect-ratio flag to stretch it back at playback time. Still-image formats have no such flag, so the extracted frame keeps the stored 1440 x 1080 geometry and looks a quarter too narrow. Set Image resolution to Width x Height and enter 1920 by 1080, and the picture comes out in the proportions you saw in the player.

Does the file need to be named .m2ts rather than .MTS?

For this page, yes — it accepts .m2ts only. The two are the same BDAV transport stream: camcorders write the short .MTS name to the card because AVCHD's folder structure follows 8.3 naming, and the clip commonly appears as .m2ts once it has been copied to a computer. Renaming the file changes nothing about its contents, or you can use MTS to HEIF, which is the same conversion with the other extension accepted.

What can actually open the .heif I get back?

In a browser, only Safari 17 and later — Chrome, Firefox, Edge and Opera display nothing, which caniuse puts at roughly 15% global support for the format. Outside the browser, macOS and iOS open HEIF natively in Photos, Preview and Quick Look, and Windows needs Microsoft's HEIF Image Extensions plus the HEVC Video Extensions installed. If the still has to open on a machine you do not control, use M2TS to JPG instead.

Is my timestamp being ignored?

It should not be, but transport streams have a specific failure mode worth knowing about. A .m2ts carries no index table, so seeking works by position and then decoding forward to the next usable keyframe. On an intact recording that is fine. On a truncated or damaged capture, a timestamp can land somewhere with nothing decodable after it and return an empty result — we reproduced exactly that on a transport stream whose only keyframe sat at the very start. If a timestamp gives you nothing, try one a little earlier, or switch to Multiple Screenshots, which reads from the beginning rather than seeking.

Is there any point choosing the Highest quality preset?

Very little. Highest maps to quality 100 and the default Very High to 95, and on the frame we measured the two files came out within a byte of one another. The rung that actually changes something is High at quality 90, which came back about 7% smaller. So treat Very High as the top of the useful range and step down if you want a smaller file, rather than reaching up for Highest.

Does the frame keep the recording date from the camcorder?

No. AVCHD stores shooting date, camera model and similar details in transport-stream and AVCHD sidecar structures that have no counterpart in a still image, and this conversion carries the picture rather than the bookkeeping. If the camera burned a date into the video itself, that text survives because it is part of the image. Otherwise, note the details from the original file before you convert.

Can I pull a whole sequence of stills in one pass?

Yes. Switch Frame Selection to Multiple Screenshots and pick a Capture Rate — the list runs from one frame every ten seconds up to ten frames per second, starting at one frame a second. Each captured moment comes back as its own separate .heif; nothing is assembled into a sequence or an animation, because HEIF as produced here is a single still image.

What happens to my recording after the conversion?

It is uploaded over an encrypted connection, decoded on our servers, and both the upload and the stills it produced are deleted automatically after a few hours. Files are never shared or made public, there is no sign-up and no watermark, and your original .m2ts is never modified — each frame is written as a new file for you to download.