To crop a video to 9:16, you select an in-frame rectangle whose width-to-height ratio equals 9 to 16 — for example, 1080 × 1920 or 720 × 1280 — and re-record that region as a new clip with all pixels outside the rectangle discarded. The crop is positioned with two offsets, X and Y, measured in source pixels from the top-left corner of the original frame, plus the rectangle's own width and height in source pixels. Cropping reframes the visible scene; it does not stretch or scale the pixels inside the rectangle, and it does not change the playback speed. A horizontal 16:9 recording can become a vertical 9:16 master for Reels, Shorts, TikTok, and Stories by trimming the left and right edges rather than the top and bottom, which is the exact operation performed by Video Cropper. Because every value is an untrimmed whole-pixel count, the math is predictable: divide the rectangle's height by 16 and multiply by 9 to get its width, then confirm both fit inside the source before processing.

What a 9:16 Crop Means in Pixels
A 9:16 rectangle has width = height × 9 / 16. The two numbers are interchangeable as long as that ratio holds, so 1080 wide and 1920 tall is just as much a 9:16 rectangle as 540 wide and 960 tall. Common whole-pixel 9:16 sizes that social platforms accept today include:
- 1080 × 1920 — the standard vertical HD resolution used by Instagram Reels, YouTube Shorts, and TikTok at full quality.
- 720 × 1280 — a smaller vertical HD size that produces noticeably smaller files and still meets every major platform's minimum resolution.
- 540 × 960 — a mid-size used when source footage is older, lower resolution, or shot on a phone with a small sensor.
- 360 × 640 — a lightweight size for low-bitrate uploads, drafts, or older playback devices.
The ratio is the contract, not a single resolution. A 9:16 rectangle can be any size as long as both numbers are whole pixels, each is at least 2, and width ÷ height equals 0.5625. Anything that meets those three rules will reframe your clip into a true 9:16 master.
Source Requirements and Size Limits
Video Cropper accepts a browser-decodable video in MP4, WebM, MOV, M4V, or Ogg. The decoded source frame must satisfy several constraints before any crop can succeed. The input shares an explicit safety budget: positive files up to 500 MiB, decoded duration above zero and no more than five minutes, no source side above 4096 pixels, and no source area above 3840 × 2160 pixels. An accepted extension does not guarantee codec support, because decoding depends entirely on the current browser.
The output rectangle inherits two extra limits of its own. Every side must be a nonnegative whole number, width and height must each be at least 2 pixels, X + width must not exceed the source width, Y + height must not exceed the source height, and the output area (width × height) cannot exceed 3840 × 2160. A 1080 × 1920 crop produces an area of 2,073,600 pixels, which sits well below the 8,294,400 ceiling. A 2160 × 3840 crop would max out the same limit at exactly the boundary. Invalid rectangles fail without shifting, clipping, rounding, or producing a partial file, so if the rectangle you want is too large for your source the operation simply refuses to start.
How to Calculate a 9:16 Rectangle Inside Your Source
Three numbers drive the calculation: source width (Sw), source height (Sh), and the target 9:16 height you want (Th). The crop width is then Tw = Th × 9 / 16, and Tw must round to a whole number that does not exceed Sw. The X offset centers the rectangle horizontally as X = floor((Sw − Tw) / 2), and Y = 0 unless you want to push the crop down.
Worked example. Source frame is 3840 × 2160 (a horizontal 4K recording). Target 9:16 height is 2160, the same as the source height, so no vertical resolution is lost. Target width = 2160 × 9 / 16 = 1215, which is a whole number and fits inside the 3840-wide source. To center the crop horizontally, X = (3840 − 1215) / 2 = 1312.5; the tool rejects fractions, so X = 1312 (floor) and the leftover half-pixel falls on the right edge. The resulting rectangle is X = 1312, Y = 0, width = 1215, height = 2160, which yields a 9:16 image with vertical resolution preserved and the source centered.
If the source is 1920 × 1080 instead, the same target height of 1080 produces a target width of 1080 × 9 / 16 = 607.5. Because the tool rejects fractions, you must choose 607 (ratio 607:1080 ≈ 0.5619, very close to 9:16 but not exact) or drop the target height to 1072 so the width becomes 1072 × 9 / 16 = 603. The rectangle cannot be 607.5. When exact 9:16 matters more than filling the full source height, plan for a slightly smaller rectangle rather than a fractional one.
Crop Your Video to 9:16 With Video Cropper
- Open Video Cropper and choose a browser-decodable video file from your device. Wait until the source's decoded dimensions appear in the form, because those numbers define the legal X, Y, width, and height range.
- Enter the four coordinates as whole pixels: X (rightward offset from the top-left), Y (downward offset), width, and height. The width:height ratio should equal 9:16, X + width must not exceed the source width, and Y + height must not exceed the source height. The defaults select the full source frame; replace them with your 9:16 rectangle.
- Select Crop video, leave the tab open while the source plays in real time, then download the resulting WebM and review it fully in your destination player before deleting the original.
Processing happens in real time, so a one-minute source takes about one minute. The output bitrate is derived from the cropped pixel area at 30 fps and is bounded between 180 kbps and 8 Mbps, which is a disclosed browser setting rather than a guaranteed file size. Fast motion, noise, grain, the source codec, and the browser encoder can all change the final size and visual quality, so review the full WebM rather than just the first few seconds.
Crop vs. Resize vs. Trim for 9:16
Choosing between three common operations matters because each one handles a 9:16 target differently. The table below compares what each operation actually does to your frames.
| Operation | What it does | Effect on pixels | Best for |
|---|---|---|---|
| Crop to 9:16 | Selects a 9:16 rectangle inside every frame and re-records it | Discards outer pixels; preserves pixel density inside the rectangle | Reframing horizontal footage for vertical platforms |
| Resize to 9:16 | Stretches or fits the entire source into a 9:16 output frame | Distorts or letterboxes the image; never throws pixels away | Matching an upload spec when source aspect already matches |
| Trim by time | Keeps the original frame size, just shortens the start and end | No re-framing at all | Removing intro or outro seconds |
A resize cannot turn horizontal footage into a vertical 9:16 master without distortion, and a trim cannot reframe anything. Cropping is the only one of the three that genuinely converts a 16:9 recording into a 9:16 deliverable while keeping the remaining pixels sharp. Use the Video Resizer when the source aspect already matches and you simply need to meet an exact byte or pixel target.
What the Crop Does and Does Not Preserve
The crop changes the visible frame region for every frame, so everything inside the rectangle is preserved at full quality and everything outside is gone forever. One fixed rectangle is applied to every frame, which means the tool does not track a moving subject or change the rectangle over time. Use exact coordinates when a platform requires a known frame or when unwanted edges are static throughout a clip; otherwise the crop will follow the static edges and the subject will drift.
Audio survives only when the browser exposes a capturable track through media capture, and the source may play audibly while the tool records, so have headphones ready and verify sound in the final file. The crop does not preserve subtitles, chapters, attachments, rotation tags, HDR signaling, color metadata, or every container feature. It changes the visible frame region; it is not a trim-by-time, resize, compress-only, redact, blur, or object-removal tool.
The output is a WebM recorded through the browser's first available MediaRecorder among VP9, VP8, and generic WebM. Because streaming MediaRecorder output may omit a finite container duration, the tool uses the same standards-tested WebM duration writer that the Video Compressor family uses: it parses Segment Info and TimestampScale and inserts a big-endian Matroska Duration value in timestamp-scale ticks, as defined in the IETF Matroska Media Container Format and rendered through the WHATWG HTML Canvas drawImage specification. Keep the original until the entire output has been reviewed, because no partial recovery is possible from a rejected rectangle or a clipped encode.