MJPEG to AVCHD Converter

What an .avchd File From This Page Actually Is

AVCHD is the consumer high-definition camcorder format Sony and Panasonic introduced in 2006, amended in 2011 to add 1080-line 50p and 60p recording and stereoscopic video. It is not really a file format at all — it is a recording specification: H.264 video and Dolby Digital (AC-3) audio, multiplexed into an MPEG transport stream, written into a BDMV directory structure on the card with the clips themselves in BDMV/STREAM as .MTS files.

This page cannot write a directory structure, because it produces one file. What it writes is the stream layer of that specification: H.264 video and AC-3 audio in an MPEG transport stream, under an .avchd filename. That is genuinely useful for feeding an editor or a player that wants AVCHD-shaped essence, and it is genuinely not an AVCHD volume you can put back on a camcorder.

The distinction is worth being precise about, because it decides what to do with the download.

AVCHD on a Camcorder vs the File This Page Writes

The practical upshot: rename the download to .ts. That extension matches what the file actually contains — a plain transport stream — and every player, editor and media server that handles transport streams will then open it without argument. Renaming it to .m2ts would be the wrong call, because that extension implies the 192-byte BDAV framing the file does not have.

How to Convert MJPEG to AVCHD

1. Upload Your MJPEG File: Drag the .mjpeg onto the page or click "Add Files" to browse. Several files can be queued and they share the same settings.
2. Choose a Compression Mode: File Compression opens on Quality Preset at "Very High (Recommended)". Constant Quality exposes a CRF slider running from 16 to 51 and opening at 23, which is the control to use when you want a predictable picture rather than a predictable size; Constant Bitrate is what a hardware target usually wants.
3. Set the Resolution and Trim if Needed: Video resolution opens on Keep original. AVCHD's own recording modes are 1920 x 1080, 1440 x 1080 and 1280 x 720, so if you are matching a camcorder, use Width x Height and enter one of those. Trim stays on Unchanged unless you switch it to Time Range.
4. Convert, Rename and Use: Click Convert, then rename the .avchd download to .ts. Files are uploaded over an encrypted connection, re-encoded on our servers and deleted automatically after a few hours — no sign-up, no watermark.

Two Things Your MJPEG Source Brings With It

A raw .mjpeg file is a run of complete JPEG pictures with no container header, and that has two consequences on this page.

There is no frame rate in the file. The decoder assumes 25 frames per second because that is the default for a JPEG sequence. A capture made at 10, 15 or 30 fps therefore plays back at the wrong speed — every frame is present, only the timing is invented — and this page has no frame-rate control to correct it. A measured example: a thirty-frame sequence captured over three seconds at 10 fps came back as a 1.200-second clip.

There is no audio, and there cannot be. MJPEG carries pictures only. So although AC-3 is the correct, specification-legal audio codec for AVCHD and this page defaults to it, the file you get will have no audio track at all. If you need sound alongside the picture, add it in an editor after conversion.

The Controls That Render on This Page

Preset Resolutions set the height and let the width follow the source's aspect ratio, so they cannot give you an exact frame — use Width x Height when a specific raster matters. Note also that leaving resolution on Keep original preserves the source's pixel aspect ratio, while any resize flattens it to square pixels.

Frequently Asked Questions

Can I copy the result back onto my camcorder's memory card?

No. A camcorder expects a complete AVCHD volume — a BDMV directory containing STREAM, CLIPINF and PLAYLIST folders with matching index and clip-information files, and stream files that use BDAV's 192-byte packet framing. This page writes one flat transport-stream file with plain 188-byte packets. Editors and players will read it happily; a camera will not see it as recorded footage.

Should the download be renamed to .ts, .mts or .m2ts?

.ts. That is the extension for a plain MPEG transport stream, which is exactly what the file is. .mts is the name camcorders give their AVCHD clips on the card and .m2ts implies BDAV framing with 192-byte packets — neither describes this file accurately, and .m2ts in particular can lead software to expect a packet structure that is not there. If you specifically need BDAV framing, use MJPEG to M2TS instead, which produces it.

Why is there only one video codec and one audio codec to choose from?

Because the AVCHD specification only permits one of each in the configuration this page targets: H.264 for video and Dolby Digital AC-3 for audio. Offering MPEG-2 or AAC here would produce a file that is not AVCHD-shaped in any meaningful sense. Pages whose target container is more permissive — MJPEG to MP4 or MJPEG to MKV — give you the full codec lists.

Is AC-3 audio actually better than AAC here?

For this target, yes, but not because it sounds better. AC-3 is one of the two audio codecs the AVCHD specification allows, alongside linear PCM, and Blu-ray players and AVCHD-aware editors are built to expect it. AAC in a transport stream is perfectly playable in software but is outside the specification, so hardware that is strict about AVCHD may refuse it. On this page the question is academic anyway, since an MJPEG source has no audio to encode.

What resolution should I use to match real AVCHD?

AVCHD's own recording modes are 1920 x 1080, 1440 x 1080 (anamorphic, 4:3 pixels displayed as 16:9) and 1280 x 720. If you are matching a camcorder's output, set Width x Height to one of those rather than using a percentage or a preset. If nothing downstream cares, leave the resolution on Keep original — upscaling an MJPEG capture to 1080p adds bytes and no detail.

Why does my converted clip run at the wrong speed?

Because the MJPEG source stores no frame rate and the decoder falls back to 25 fps. The ratio between your real capture rate and 25 is exactly the speed error, so a 30 fps capture runs about 20% slow and a 10 fps capture runs 2.5 times fast. All the frames are there. The only reliable fix is to re-wrap the JPEG sequence at its true rate in a desktop tool before uploading; nothing on this page can override the assumption.

Will the picture quality be better than the MJPEG I started with?

No — it cannot be. The server always re-encodes: your JPEG frames are decoded and compressed again with H.264, so this is a second lossy generation. What improves is efficiency, not fidelity. H.264 uses motion prediction between frames, which Motion JPEG does not do at all, so the same picture typically fits in a small fraction of the bytes. Keep the CRF at or below 23 if you want the loss to be hard to see.

Why is my transport-stream output bigger than I expected for such a short clip?

Transport streams are designed for broadcast, not for storage efficiency. Every 188-byte packet spends four bytes on a header, program tables are repeated periodically so a receiver tuning in mid-stream can sync, and null packets may be inserted to hold a constant rate. That overhead is a few per cent and it never goes away. If size matters more than transport-stream compatibility, MJPEG to MP4 carries the same H.264 video in a much leaner container.

How are my files handled during conversion?

Your .mjpeg is uploaded over an encrypted connection, decoded and re-encoded on our servers, and both the upload and the resulting file are deleted automatically after a few hours. Nothing is shared or made public, and there is no sign-up, no watermark and no account. Motion JPEG is bulky for its running time because nothing is shared between frames, so if you only need part of the capture, use the Trim control rather than uploading the whole thing twice.