HN Showcase
Back to browse
A segmentation model client-side via WASM – free background removal

A segmentation model client-side via WASM – free background removal

by shivaodin·Feb 12, 2026·4 points·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidEye CandySolve My Problem

Fully client-side ML inference; 6 tools share one cached model—but remove.bg and Cleanup.pictures already own this.

Strengths
  • •Zero-upload architecture: image processed locally in 2-3 seconds, cached model speeds future runs
  • •One segmentation model powers six derivative tools (changer, passport, whitener, blur, sticker)—elegant reuse
  • •Offline capable after first ~40MB download; genuinely private, no tracking
Weaknesses
  • •Crowded market: remove.bg, Cleanup.pictures, Canva all do background removal instantly and have better edge quality
  • •WebGPU support is still inconsistent; WASM fallback may be slow on older devices
Category
Design
Target Audience

Content creators, e-commerce, product photographers, social media users

Similar To

remove.bg · Cleanup.pictures · Canva background remover

Post Description

Built a background removal tool that loads a ~40MB segmentation model into the browser via WASM/WebGPU and runs inference client-side.

No upload step, no API call, no queue. Drop an image, get the result in 2-3 seconds. No per-image charges because there's no server doing the work.

The same cached model powers 6 derivative tools — background changer, passport photo maker, product photo whitener, portrait blur, sticker maker — each just different post-processing on the same mask output.

Similar Projects