Initializing... drag & drop files here
Supports: MOV
A .mov is video; a .jpg is a single still. So this is not really a format conversion — it is frame extraction, and the only decision that matters is which frame (or frames) you want out. This page pulls a frame at a timestamp you choose, or a whole sequence at a fixed capture rate, and encodes each one as a JPEG at the quality you pick.
.mov onto the page or click "+ Add Files". Several clips can be queued and processed in one batch with the same settings.| Mode | Control | What you get | Use it for |
|---|---|---|---|
| Specific Frame (opens selected) | Time (seconds) — starts at 0 | One JPG, taken at that timestamp | Thumbnails, a poster frame, one specific moment |
| Multiple Screenshots | Capture Rate — one frame every 0.1 s (10 fps) through to one every 10 s | One JPG per captured frame | Contact sheets, frame-by-frame review, storyboards, training data |
Leave Time (seconds) at 0 and you get the literal first frame of the file — which on real footage is very often the worst frame in the clip. Camera and phone recordings frequently open on a moment where autofocus and auto-exposure are still settling, edited exports usually start on a fade from black, and screen recordings tend to begin on the instant before anything happened. Scrub the clip in QuickTime or VLC, note the second you actually want, and type that number in. Fractional values work, so 4.5 is a valid timestamp.
The Capture Rate values are expressed as an interval, not a frame rate: "0.1s seconds (single frame at 10fps)" is the fastest, "10 seconds per frame" the slowest. Multiply before you commit — a two-minute clip at one frame every 0.1 s is roughly 1,200 JPEGs, while the same clip at one frame per second is 120. If you only need a section, trim the source first with Video Cutter and extract from the shortened clip.
The frame is pulled out of the video as a lossless PNG first and only then encoded to JPEG, so there is exactly one lossy generation between the video frame and your file — nothing is re-compressed twice. The preset maps to a plain JPEG quality value:
| Quality Preset | JPEG quality | Sensible use |
|---|---|---|
| Highest | 100 | Archival stills, frames you will edit further |
| Very High (opens selected) | 95 | Default; visually indistinguishable at normal viewing sizes |
| High | 90 | Web-ready with a useful size drop |
| Medium | 80 | Thumbnails, contact sheets, bulk sequences |
| Low | 70 | Small previews where artefacts are acceptable |
| Very Low | 60 | Aggressive size reduction |
| Lowest | 50 | Smallest file, clearly visible JPEG blocking |
If a hard byte budget matters more than a quality tier, switch Image Compression to Specific file size and give a number in KB or MB; the encoder works backwards to that target. And if you are pulling a sequence, remember the quality choice multiplies — the gap between Very High and Medium on one image is trivial, but across 1,200 frames it is the difference between a manageable download and an unmanageable one.
JPEG is an 8-bit, chroma-subsampled, lossy still format from 1992. It holds one image, so anything about the source that is not "these pixels at this instant" is gone: audio, motion, duration, chapter markers and QuickTime metadata atoms all stop at the frame boundary. It also has no alpha channel, so a ProRes 4444 MOV with transparency lands as opaque pixels.
Two source-side caveats are worth knowing before you blame the output. HDR footage from a recent iPhone — Dolby Vision or HLG — gets a plain colour conversion, because there is no HDR tone-mapping stage in the pipeline; expect flat, grey or washed-out stills unless you export an SDR version first. And interlaced footage from an older camcorder is not deinterlaced anywhere, so a fast-motion frame will show combing lines that were always in the source field pair. For a pixel-exact still with no JPEG artefacts and full alpha support, use MOV to PNG instead; for a moving result, MOV to GIF.
The first frame in the file. That default exists so the tool does something sensible without configuration, but it is rarely the frame you want from real footage — clips commonly open on a fade, a focus hunt or an exposure ramp. Set Time (seconds) to the moment you actually care about; decimals such as 12.75 are accepted.
Play the clip in QuickTime, VLC or your editor, read the timecode at the moment you want, convert it to seconds, and enter that in Time (seconds) with Specific Frame selected. The extractor seeks to that position and takes one frame. If you need several precise moments rather than a regular interval, run the conversion once per timestamp — it is faster than pulling a sequence and hunting through it.
Clip length divided by the capture interval. The dropdown is written as an interval, so "1 second per frame" on a 3-minute clip is about 180 images, and "0.1s seconds (single frame at 10fps)" on the same clip is about 1,800. There is no frame-count cap in the tool, so do the multiplication first — the practical limit you will hit is upload size and download volume, not the extractor.
Only from the JPEG encode itself. The frame is extracted as a lossless PNG and converted once, so there is no double-generation loss stacking on top of the video's own compression. At the Highest or Very High presets (JPEG quality 100 and 95) the result is visually indistinguishable from the decoded frame. What you cannot recover is detail the video codec already discarded — a heavily compressed 720p MOV yields a heavily compressed-looking still no matter what quality you pick.
Use JPG for thumbnails, previews, sequences and anything going onto a web page, where the file-size advantage is large and the artefacts are invisible at normal sizes. Use PNG when the frame contains text, UI elements or sharp graphic edges — screen recordings are the obvious case — because JPEG's chroma subsampling smears coloured text, and when you intend to edit and re-save the image repeatedly.
Whatever the video's frame size is, if you leave Image resolution on Keep original — a 4K MOV yields a 3840 × 2160 JPEG. Preset Resolutions sets the height and lets the width follow your source's aspect ratio, so a 16:9 clip at the 720p preset gives 1280 × 720. Use Width x Height only when you want an exact pixel box and accept that it may not match the source aspect.
Because there is no HDR tone-mapping step in the conversion. Dolby Vision and HLG footage carries brightness and colour information that an 8-bit JPEG cannot represent, and a straight conversion flattens the highlights and desaturates the image rather than remapping it. Export an SDR copy from Photos or your editor first, then extract the frame from that.
.jpeg extension instead of .jpg?Yes. The File extension group offers both JPG and JPEG; neither is preselected, so pick the one your workflow expects. The bytes are identical — it is the same JFIF-wrapped JPEG data either way — so this only matters when a downstream system does a strict filename check.
It is uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours. Nothing is shared, published or made searchable, there is no sign-up, and no watermark is added to the extracted frames.