Choose a photo
The selected photo is read directly in your browser instead of being sent to a Siamese Cat image-processing API.
Upload a photo, adjust the effect in real time, and download the result. A small AI model helps locate the torso while the selected image stays in your browser.
JPG, PNG or WebP up to 20 MB
You can also drag and drop a photo here.
A lightweight pose model estimates shoulder and hip landmarks, which helps place the waist focus. The actual visual transformation is a real-time WebGL pixel warp rather than a large generative image model.
The selected photo is read directly in your browser instead of being sent to a Siamese Cat image-processing API.
Pose Landmarker estimates the body focus. If detection is unclear, simply tap the photo to place the focus yourself.
Fine-tune strength, effect area and blend, compare with the original, then save the finished image as a PNG.
The selected image is not sent to a Siamese Cat server by this tool. The MediaPipe runtime and model are downloaded to the browser and inference happens on the user's device.
WebGL handles the visual warp, so slider changes are rendered immediately without waiting for a server-side processing queue.
If AI detection is unavailable or turned off, the user can tap the image to place the reshape focus manually and continue editing.
This free tool connects the wider Siamese Cat ecosystem — Siamese Cat Creative Club, Siamese Cat Cafe, and the product / development work of .
No. The tool reads and edits the selected image in your browser. It does not send that photo to a Siamese Cat server. The page does download the MediaPipe runtime and AI model from a CDN/Google Storage so the AI can run on the device.
Yes, but only a small pose-detection model to identify useful body landmarks such as the shoulders and hips. The actual reshape is a WebGL pixel warp, not a large generative image model.
This type of reshape moves pixels around the torso. Strong settings can affect nearby straight background lines. Lower the strength or effect area for a more natural result.
It is designed for modern browsers with JavaScript and WebGL. AI speed depends on the device and browser. If AI detection cannot load, the manual focus mode still works.
No. It only changes the appearance of a photo for creative editing. It does not measure weight, health, body composition, or a person's real-world proportions.
Use photos you own or have permission to edit. This is an adult portrait editor: do not use it on children’s photos. Your selected image stays in your browser; optional pose detection is processed on your device.