Back to browse
GitHub Repository

LiveLook — Real-time virtual try-on with gesture control, powered by Decart's Lucy VTON model

5 starsTypeScript

Real-time virtual try-on using hand gestures and live video diffusion

by manas95·May 21, 2026·10 points·1 comment

AI Analysis

MidEye CandyShip It

Gesture-controlled try-on wrapper around Decart's Lucy VTON when Shopify already does this.

Strengths
  • MediaPipe gesture recognition adds natural browsing without buttons.
  • WebRTC streaming keeps latency low for live video diffusion.
  • Voice search via Web Speech API reduces typing friction.
Weaknesses
  • Zero stars, zero forks — very early stage with no traction.
  • Core tech entirely from Decart and Google; no novel architecture.
Category
Target Audience

E-commerce developers, retail tech teams

Similar To

Vue.ai · Revery.ai · Zyler

Similar Projects

AI/ML●●Solid

Browser-based hand gesture T9 keyboard (YOLOX and ONNX Runtime Web)

Runs everything in the browser and actually stays responsive — ONNX Runtime Web + a YOLOX model handle subtle hand-seal distinctions that MediaPipe struggled with. Clever choice to layer a T9 keypad over gesture input (reduces required class count and makes errors tolerable), but the demo remains an experiment: lighting sensitivity and similar seals create real UX friction and it’s not yet a drop-in input alternative.

WizardryNiche Gem
huang4fun
203mo ago
AI/ML●●Solid

MirrAI – I built a Virtual Try-On app to fix online shopping(Open Beta)

The app splits the difference between fun and useful: a 'Fit Check' flow that attempts to preserve your pose and lighting when blending garments feels more practical than a pure novelty image generator, and there's a text-to-image 'Imagine' mode plus a community feed. The implementation looks thoughtfully productized for mobile, but the obvious limits — fabric physics, glitches on complex poses, and unclear privacy/compute tradeoffs — keep it from feeling like a step-change.

Crowd PleaserSlickShip It
MohammadRMZ
103mo ago