Track, Mask, Cut Out, or Blur Any Object in Video

Yes. AI Video Masking is completely free, requires no signup, and adds no watermark to your exported video.

Upload an MP4 or MOV, click the object you want to track on the first frame, pick your outputs, and hit Track Object. The model propagates the selection through every frame and exports everything you selected in one run: transparent cutout, blurred video, black & white mask, or white alpha matte.

Yes. Instead of drawing a mask frame by frame, click the subject once and SAM2 automatically propagates the mask through the video. You can then export the tracked result as a transparent cutout, blur, black & white matte, or white alpha matte.

Foreground clicks (green dots) tell the model which pixels you want to keep. Background clicks (red dots, added by right-clicking or switching to background mode) tell it which pixels to exclude. Adding a background click near an unwanted region can help refine a noisy selection. The selection preview updates instantly after every click.

The tool accepts MP4, MOV, and WebM files up to 600 MB. Shorter, lower-resolution clips process faster and are more reliable on modest GPUs.

Not currently. Mobile browsers are disabled because this tool depends on desktop WebGPU performance. Use an up-to-date desktop version of Chrome or Edge.

Meta's Segment Anything 2.1 video model — the same architecture behind state-of-the-art video object segmentation. Unlike image-only ports, FuzzPuppy runs the complete video pipeline including SAM2's memory attention: the model remembers what your object looks like across recent frames, follows it through motion, and recovers it after occlusions. Everything runs on your GPU via WebGPU.

No. All tracking and encoding runs locally in your browser. Your source video never leaves your device.

After the first download (~190 MB), the model is cached in your browser's storage so the tool starts instantly on later visits. Nothing is installed on your computer, and your browser clears the cache automatically if disk space runs low. To remove it manually, use the 'Delete downloaded model' button at the bottom of the tool, or your browser's Clear Site Data option.

The first run downloads the model (about 190 MB) — it's cached by your browser, so later runs skip the download. After that, every frame is processed locally on your GPU. Speed depends on your GPU, video length, and resolution. Shorter or lower-resolution clips process noticeably faster.

Browsers throttle background tabs. Keep the tracker tab visible while it processes so frame extraction and AI inference run at full speed.

Four, in any combination from a single tracking run: an object cutout MOV (just the tracked object on a transparent background, with audio), a blurred MP4 (the full video with the object blurred — or the background blurred for a focus effect — with audio), a black & white mask MP4 (for editor luma/track mattes and AI video models that accept mask inputs), and white alpha matte MOVs (the object's white shape on transparency — standard, inverted, or both).

To place the object onto other footage, use the object cutout MOV. To hide a face, license plate, or logo — or to fake a shallow depth-of-field — use the blurred MP4. For full control, use the black & white mask as a luma matte / track matte in Premiere Pro, After Effects, DaVinci Resolve, Final Cut Pro, or CapCut. You can also use it as a mask input for compatible AI video models and workflows, including masked generation, inpainting, object replacement, and region-specific effects. The white alpha matte is for stylized glow, outline, and stencil-reveal effects.

The MOVs use a lossless alpha codec that QuickTime Player previews poorly. They work correctly in editors (CapCut, Premiere Pro, After Effects, DaVinci Resolve, Blender). To make a ProRes 4444 copy for QuickTime playback, run:

The object cutout MOV and the blurred MP4 keep the source audio. The mask and white alpha matte outputs are video-only — they're mattes, and your original clip keeps its audio on another track in your editor.

Try adding a few more foreground clicks on the object and a background click to rule out nearby distractors before hitting Track. Clips where the object moves smoothly and stays clearly visible produce the best results. Very fast motion, heavy motion blur, and long full occlusions are the hardest cases — though the tracker's memory lets it recover the object after it reappears.

WebGPU is available in up-to-date desktop Chrome and Edge. Open WebGPU Report to confirm your browser detects a WebGPU adapter. If it doesn't, update your browser and GPU drivers, and make sure hardware acceleration is enabled in your browser settings.

Yes. If the object is fully hidden or leaves the frame, those frames are exported as fully black and the tracker automatically reacquires the object when it reappears nearby. The tool reports how many frames the object was hidden after processing.