[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-94341":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":8,"htmlUrl":8,"language":9,"languages":8,"totalLinesOfCode":8,"stars":10,"forks":11,"watchers":12,"openIssues":13,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":14,"stars7d":14,"stars30d":15,"stars90d":14,"forks30d":14,"starsTrendScore":14,"compositeScore":16,"rankGlobal":8,"rankLanguage":8,"license":8,"archived":17,"fork":17,"defaultBranch":18,"hasWiki":17,"hasPages":19,"topics":20,"createdAt":8,"pushedAt":8,"updatedAt":21,"readmeContent":22,"aiSummary":23,"trendingCount":14,"starSnapshotCount":14,"syncStatus":24,"lastSyncTime":25,"discoverSource":26},94341,"finger-frame-effect-lucy","sophiamyang\u002Ffinger-frame-effect-lucy","sophiamyang",null,"JavaScript",197,49,3,1,0,24,41.5,false,"main",true,[],"2026-08-24 04:01:22","# Finger Frame Live AI 🎬⚡\n\n**Try it: https:\u002F\u002Fsophiamyang.github.io\u002Ffinger-frame-effect-lucy\u002F**\n\nHold up both hands and frame a box with your fingers — and see a **live,\nreal-time AI world inside the frame**, generated at 30fps by\n[Decart Lucy 2.5](https:\u002F\u002Fdocs.platform.decart.ai\u002Fmodels\u002Frealtime\u002Flucy-2.5)\n(realtime video-to-video over WebRTC). Unlike offline generation, the AI\nversion moves *with* you: blink and it blinks, wave and it waves — with tens\nof milliseconds of model latency.\n\n## The finger-frame family\n\n| App | Generation | Latency |\n|---|---|---|\n| [finger-frame-effect](https:\u002F\u002Fsophiamyang.github.io\u002Ffinger-frame-effect\u002F) ([repo](https:\u002F\u002Fgithub.com\u002Fsophiamyang\u002Ffinger-frame-effect)) — live camera, local effects | Canvas 2D (Van Gogh, toon, glitch, …) | none |\n| [finger-frame-effect-ai](https:\u002F\u002Fsophiamyang.github.io\u002Ffinger-frame-effect-ai\u002F) ([repo](https:\u002F\u002Fgithub.com\u002Fsophiamyang\u002Ffinger-frame-effect-ai)) — recorded video, AI restyle | Gemini Omni Flash (offline video edit) | minutes |\n| **this app** — live camera, live AI | Decart Lucy 2.5 (realtime video-to-video) | ~real time |\n\n![Example: live AI world inside the finger frame](examples\u002Flucy.gif)\n\n*Live capture — the AI world inside the frame moves in real time with the\ncamera ([full-quality mp4](examples\u002Flucy.mp4)).*\n\n## How it works\n\n- The webcam feed (1280×720@30) is mirrored to a full-screen canvas, and\n  MediaPipe Hand Landmarker tracks the finger-frame quad with the pipeline\n  from the original app (anatomical corner ordering, gesture hysteresis,\n  teleport rejection, velocity-adaptive smoothing, dropout hold).\n- The same camera stream is sent to Lucy 2.5 over WebRTC; the transformed\n  stream comes back live and is drawn screen-aligned inside the quad — the\n  finger frame is a window into the AI world.\n- Effects (keys 1–6) are **live style prompts**: switching sends the new\n  prompt into the running session with no reconnect. Custom ✨ uses your own\n  prompt from the 🔑 panel.\n\n## Bring your own key\n\nGet a key at [platform.decart.ai](https:\u002F\u002Fplatform.decart.ai\u002F), paste it in\nthe 🔑 panel. It stays in your browser and is used only for the WebRTC\nsession with Decart. Without a key, the window falls back to a local color\nfilter so the tracking is still demoable.\n\n## Run locally\n\n```bash\npython3 -m http.server 8125\n```\n\nOpen http:\u002F\u002Flocalhost:8125 and allow camera access. `?demo` runs a synthetic\nfeed with fake hands for testing the tracking without a camera (AI disabled\nin demo mode).\n","这是一个基于浏览器的实时AI视频特效项目，通过手指框选手势在摄像头画面上创建动态窗口，并实时调用Decart Lucy 2.5模型进行视频到视频转换，实现低延迟（约毫秒级）的AI生成内容随用户动作同步响应。核心依赖MediaPipe手部关键点追踪与WebRTC流式AI推理，支持动态切换风格提示词，无需重连即可实时生效。适用于交互式AI艺术展示、教育演示、创意媒体实验等需要轻量级、免安装、手势驱动实时AI视觉反馈的场景。",2,"2026-08-07 02:30:03","CREATED_QUERY"]