HUANGCHIHHUNGLeo / claude-real-video

Let Claude (or any LLM) actually watch a video — scene-aware, deduplicated frames + transcript, from a URL or local file. Runs locally, MIT.

xiaohui5206 / let-ai-read-video

让任何 AI 真正"看"视频：纯本地的视频阅读 skill——faster-whisper GPU 转写 + ffmpeg 场景感知抽帧，双通道带时间戳（t=MM:SS），数据不出机。1 小时视频 3~5 分钟读完。Give any AI agent eyes for video: 100% local video-reading skill — faster-whisper GPU transcription + scene-aware ffmpeg frame extraction, dual-channel with t=MM:SS citations. Nothing leaves your machine.

morishuz / adaptive-frame-extractor

Adaptive video frame extraction for SfM, Gaussian Splatting, and photogrammetry.

yuxino-labs / koma

Self-hosted AI video analysis: summaries, chapters, keyframes, transcripts and SRT / JSON export. AI 视频总结、字幕提取与结构化分析。