一款通过摄像头和MediaPipe实现手势控制的3×3拼图游戏,用户可以用捏合手势框选场景、拖动拼块,最终以电影胶片条形式保存。

Stars

41

7 天增长

暂无数据

Fork 数

0

开放 Issue

0

开源协议

暂无数据

最近更新

2026-07-05

AI 仓库情报摘要
FR-AI / ANALYSIS

为什么值得关注

它将实时手部追踪与经典拼图机制融合在一个网页应用中,依靠MediaPipe和原生JavaScript实现,并配有独特的胶片条输出功能。

适合谁使用

  • 探索手势交互的Web开发者
  • 寻求新颖互动方式的拼图爱好者
  • 对MediaPipe和摄像头项目感兴趣的创意编程者
  • 想要趣味性免触控照片编辑的用户

典型使用场景

  • 具有手势控制的互动艺术装置
  • 用于手部追踪与计算机视觉的教学演示
  • 利用摄像头互动的社交或派对游戏
  • 基于手势的照片编辑与拼贴原型

项目优势

  • 借助MediaPipe HandLandmarker实现精确的手部追踪
  • 使用原生JavaScript,无框架依赖,轻量且易于本地运行
  • 新颖的交互流程(框选、拍摄、解谜、保存),视觉反馈清晰
  • 最多容纳3张照片的胶片条画廊,支持PNG下载

使用前须知

  • 需要WebGL和摄像头支持,仅能在Chrome或Edge浏览器中运行
  • 只支持固定的3×3拼图尺寸
  • 每张胶片条最多3次拍摄,重置前无法继续添加

README 快速开始

Frame Puzzle

A hand-gesture-controlled puzzle game using your webcam and MediaPipe.

Frame a scene with both hands (pinch), solve the 3×3 puzzle by dragging pieces with pinch gestures, then fist to save. Captures build up in a film strip gallery! download the strip when it's full.

How it works

  1. Frame — Pinch with both hands to define a capture area. Hold for 3 seconds.
  2. Capture — A photobooth-style snapshot (B&W, high contrast, film grain) is taken.
  3. Solve — The image is shuffled into 9 tiles. Pinch a tile and drag it to its correct position.
  4. Save — Make a fist to save the completed puzzle. The tiles shatter and the result lands in the film strip.
  5. Strip — Up to 3 captures per strip. Download the full strip as PNG or reset.

Run locally

npx serve .

Open the URL shown in your terminal. Chrome or Edge recommended (requires WebGL for MediaPipe).

Tech

made by talha

相关仓库与替代方案

根据分类、Topic 和编程语言匹配的相似项目。

react
精选
react GitHub avatar

react

React is a JavaScript library for building user interfaces with declarative, component-based, and cross-platform design.

Web 开发前端框架
246,741
tandpfun
精选
tandpfun GitHub avatar

wardrobe

A GPT‑image powered app that extracts individual garment cutouts from photos, organizes them into a digital wardrobe, and generates modeled outfit previews.

JavaScript
1,570
mshumer
精选
mshumer GitHub avatar

Claude-of-Duty

A browser-based first-person shooter built entirely with procedural generation and orchestrated AI agents, featuring 55k lines of Three.js/WebGL2 code and no art assets.

AI 与机器学习AI 智能体
1,234