AVCHD to BMP Converter

Grab a Still Frame From AVCHD Camcorder Footage as BMP

This is a frame grabber, not a video converter: an .avchd clip goes in, one or more Windows bitmap stills come out. The conversion decodes the frame you ask for, writes it as a lossless PNG internally, then encodes that to BMP — so nothing is lost between the camcorder and the bitmap except what the BMP quality preset gives away. That preset is the one control worth understanding before you start, because on this page it chooses a colour depth rather than a compression level.

How to Convert AVCHD to BMP

1. Upload Your AVCHD File: Drag the .avchd clip onto the page or click "+ Add Files". Several clips can be queued, and each produces its own still.
2. Choose the Frame You Want: Open Advanced Options and use Frame Selection. Leave it alone to get the first frame of the clip, choose Specific Frame and enter a Time in seconds to grab a moment part-way in, or choose Multiple Screenshots and pick a Capture Rate to get a whole series of stills from one clip.
3. Pick a Quality Preset and Size (Optional): Image Compression offers High, Medium and Low — see the table below for what each one actually writes. Image resolution starts on Keep original; the preset list runs from 4320p down to 16P, and Width × Height or Resolution Percentage let you scale the still.
4. Convert and Download: Click Convert and download the .bmp. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark.

What Each BMP Quality Preset Actually Writes

BMP has no lossy mode, so a "quality" setting has nowhere to go except colour depth and run-length compression. That is exactly what these three presets change, and the difference between them is far larger than the word "quality" suggests.

The High figure is arithmetic and will hold for any frame at that size, because the file is uncompressed. The Medium and Low figures are from one test frame — RLE compresses flat areas well and busy footage badly, so your numbers will differ.

Frame Selection in Practice

Whichever you choose, the frame is decoded at full precision and written to a lossless PNG before the BMP encode runs, so nothing is lost between the camcorder stream and the bitmap except what the preset in the previous table gives away.

Frequently Asked Questions

Which frame do I get if I do not touch Frame Selection?

The first frame of the clip. That is the fast path and it is usually what you want for a thumbnail, but camcorder footage often opens on a fraction of a second of autofocus hunting or a lens cap coming away. If the still looks soft or dark, switch Frame Selection to Specific Frame and enter a Time a second or two in — the frame is decoded properly at that point rather than being approximated, so a value like 2 or 5.5 gives a clean grab.

How do I pull a whole sequence of stills instead of one?

Choose Multiple Screenshots under Frame Selection and pick a Capture Rate. The fastest setting captures a frame every 0.1 seconds (ten per second) and the slowest one every ten seconds, with several steps in between. Each captured frame becomes its own BMP, so a one-minute clip at one frame per second yields sixty files. Bear in mind the file-size table above before doing this at the High preset — sixty uncompressed 1920×1080 bitmaps is about 249 MB.

Why does the High preset still only give 16-bit colour?

Because that is what this page's High preset is wired to produce: the frame is written with five bits each of red, green and blue, which is 32,768 possible colours rather than the 16.7 million a 24-bit BMP would carry. There is no 24-bit or 32-bit option on this page. In practice 16-bit is fine for viewing and for pasting into documents, and the only place it shows is smooth gradients — a clear sky or a studio backdrop can pick up faint banding. If you need full 24-bit colour from the same frame, use AVCHD to PNG instead, which is lossless and true colour.

Why is a single still bigger than several seconds of video?

Because video codecs describe change and BMP describes pixels. AVCHD stores one full frame and then a long run of differences, so the per-frame cost of a moving picture is a few tens of kilobytes. A BMP stores every pixel independently with no inter-frame trickery and, at the High preset, no compression at all — two bytes per pixel, every time. A 1920×1080 still at High is 4,147,254 bytes no matter what is in the picture. Switching to Medium brings palette reduction and RLE into play and typically cuts that by an order of magnitude.

My still has comb lines through anything that was moving — why?

Because your source is interlaced and there is no deinterlacing step in this conversion. AVCHD camcorders often record 1080i, which means each frame is really two fields captured a fiftieth or sixtieth of a second apart. Freeze one and anything that moved between the two fields shows as horizontal comb teeth. Nothing on this page removes them. The fix is to deinterlace before converting — Handbrake's Decomb filter or ffmpeg's yadif both work — or to pick a moment where the subject was still.

My widescreen camcorder frame comes out squashed — how do I fix it?

Enter an explicit Width × Height. Some AVCHD clips store a 1440×1080 frame that is meant to be displayed as 1920×1080, using non-square pixels to stretch it at playback time. A still image has no way to carry that instruction, so the extracted bitmap is the stored 1440×1080 and looks horizontally compressed. Setting Width × Height to 1920 × 1080 in step 3 resamples it to the shape you actually saw on the camcorder screen.

Does the still keep the recording date or any camera metadata?

No. BMP has no standard metadata block — there is nowhere in the format to put a capture date, camera model or exposure settings, so none of it travels. On top of that the Medium preset strips profiles explicitly. If you need the shooting metadata alongside the picture, note it down from the clip before converting, or use a target that can carry metadata such as AVCHD to JPG.

Is BMP the right choice at all for a camcorder frame?

Only for specific reasons. BMP is uncompressed or lightly RLE-compressed and carries no metadata, which makes it a poor archive format and a bad choice for sharing. It is the right pick when something downstream demands it: older Windows applications, some embedded displays and digital signage players, certain lab and machine-vision tools, and Windows wallpaper workflows all take BMP and nothing else. For everything else, PNG gives you lossless 24-bit colour in a fraction of the space, and JPG gives you a shareable file.

Can I crop the still, or does it always cover the whole frame?

The conversion always writes the full frame — there is no crop control here, only scaling. Grab the still first, then crop it in an image editor, or use one of our image tools on the resulting file. If what you actually want is a small region blown up, extract at Keep original resolution so you have the maximum pixels to crop into rather than scaling down first. For the full list of still and video targets from the same footage, see the AVCHD converter hub.