MJPEG Compressor

Compress MJPEG Video Online

Motion JPEG stores every frame as its own full JPEG image, so raw .mjpeg recordings from IP cameras, dashcams, and webcams run enormous. This tool re-encodes the stream to a smaller MJPEG by lowering per-frame quality, dropping the resolution, thinning the framerate, or trimming length — no sign-up, no watermark. For the biggest reduction, though, the honest answer is usually to change codecs entirely: converting to H.264/MP4 shrinks the same footage far more than any in-place MJPEG setting can, because MJPEG never compresses across frames.

How to Compress an MJPEG File

1. Upload Your MJPEG File: Drag and drop your .mjpeg clip or click "Add Files". The tool accepts Motion JPEG streams from IP and security cameras, dashcams, webcams, and older digicams; drop in several at once and each is processed in parallel.
2. Pick a Compression Mode: Under File Compression, choose how you want to hit your size target — Quality Preset (the default is "Very High"), Specific file size to cap the output at an exact MB value, Target file size (%) to scale relative to the source, Constant or Variable Bitrate for predictable streaming sizes, or the CRF quality slider (2 keeps the most detail, 31 shrinks the most).
3. Resize, Trim, or Drop Frames (Optional): Under Video resolution, keep the original, pick a preset (1080p / 768p / 720p / 480p), scale By Percentage, or set a custom Width × Height with aspect locked. Use Trim to keep only a time range, or Drop Frames to thin the framerate — both cut size hard on long, static surveillance footage.
4. Compress and Download: Click Compress. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark, never shared.

MJPEG Compression Controls and What They Do

Why MJPEG Files Are So Large

MJPEG (Motion JPEG) compresses each frame independently as a baseline JPEG (ITU-T T.81 / ISO/IEC 10918-1) and never compares one frame to the next. That intra-frame-only design makes single-frame seeking trivial and keeps decoding cheap on low-power camera hardware — but it also means a full image is stored for every frame even when almost nothing in the scene changed. Motion JPEG typically reaches only a 10:1 to 20:1 compression ratio, while modern inter-frame codecs such as H.264 routinely reach 50:1 or better by storing just what changed between frames.

That gap is the key thing to understand before you compress. Staying in MJPEG, this tool re-encodes the video — it does not simply repackage it, so the output is never pixel-identical to the source, and the realistic savings come from accepting lower per-frame quality, fewer pixels, or fewer frames. It helps, but there is a floor: every frame is still a standalone JPEG. If the goal is the smallest possible file that still plays everywhere, the bigger lever is switching codecs — re-encode to H.264/MP4 and the inter-frame compression does what no MJPEG setting can. Motion JPEG's transport was only ever formalized for streaming, as an RTP payload in RFC 2435; there is no single universal file spec, which is also why an MJPEG clip that opens in one player can be rejected by another.

Compress MJPEG In Place vs. Convert to H.264

Frequently Asked Questions

How much smaller can I actually make an MJPEG file?

It depends on the footage and the settings, but in-place MJPEG compression has a hard ceiling because every frame is still a full JPEG. In our testing, a 10-second 1080p MJPEG clip of a mostly static scene started at about 78 MB; downscaling it to 720p at a lower quality setting brought it down to a fraction of that, while converting the same clip to H.264/MP4 at the Very High preset produced roughly a 4 MB file — around a 95% reduction. So compressing in place gets you a meaningful cut, but if you need the smallest file, changing codecs saves far more.

Does compressing an MJPEG re-encode it, or just repackage it?

It re-encodes it. There is no lossless "rewrap" path for making an MJPEG smaller — lowering the quality, resolution, or framerate means each frame's JPEG is decoded and re-compressed, so the output is never pixel-identical to the source. The one lossless operation is not compression at all but frame extraction: because each frame is already a JPEG, pulling stills out with MJPEG to JPG copies the picture data without recompressing it. Keep your original if you need a pristine master for archival or forensic work.

Should I compress the MJPEG or convert it to MP4 instead?

For most people, convert. MJPEG's intra-frame design is exactly why the files are big, and no in-place setting escapes that — every frame remains a standalone JPEG. Re-encoding to H.264/MP4 applies inter-frame compression that typically shrinks the file far more while keeping it playable on virtually every device and browser. Stay in MJPEG only when you specifically need per-frame, frame-accurate footage for editing, machine vision, or a workflow that requires the format.

What CRF or quality setting should I use?

The CRF slider for Motion JPEG is really a per-frame JPEG quantizer that runs from 2 (highest quality, largest file) to 31 (lowest quality, smallest file). Values around 2–5 stay visually clean and are a good choice for footage you may analyze later, such as a license plate or a face; 8–15 is a reasonable middle ground for general viewing; going much past 20 introduces visible blocking on detailed scenes. If you would rather not tune it directly, the Quality Preset dropdown maps the same idea to named steps from Highest to Lowest.

Will compressing degrade the quality?

Yes, some — because it re-encodes. How much is under your control: the Quality Preset, the CRF slider, and the output resolution each trade detail for size. At a high preset the loss is usually invisible in normal viewing; pushing for an aggressive Specific file size on detailed footage will start to show JPEG blocking. If you need to preserve every pixel, keep the original MJPEG and work on a copy.

Does my MJPEG have audio, and is it kept?

A raw .mjpeg stream is video-only — the format is a sequence of JPEG frames and carries no audio track of its own, which is the norm for IP-camera and dashcam exports. If your file was muxed into a container (some camcorders and capture cards do this) and does carry a soundtrack, the tool re-encodes it alongside the video using the selected audio codec. If there is no audio stream to begin with, there is simply nothing to keep.

Can I trim a long recording before compressing?

Yes. Surveillance and dashcam captures are often hours long, and trimming is the single biggest size lever on that kind of footage. Use the Trim control here to keep only a time range, or, for a dedicated interface, cut the clip first with the Video Cutter and then compress the shorter result. Combined with a resolution downscale, trimming turns a multi-gigabyte recording into something you can actually share.

Is there a file size limit, and is the upload private?

Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark, never shared or made public. There is no fixed per-file cap; the practical limit is upload size and your connection speed, and because MJPEG runs large, a long capture can take a while to transfer. If a multi-gigabyte file is slow to upload, trim it first, or shrink an already-converted clip with the Video Compressor.