Startup Ideas Bank
Innovative Video-Reading for AI, but Can It Generate Revenue?
AI roast score: 68/100 (C)
The idea
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.
claude-real-video
Let Claude — or any LLM — actually watch a video.
Same 58-second clip: fixed 1 fps sampling = 58 frames . crv keeps the 26 that actually differ — and --grid packs them into 3 contact sheets . Fewer tokens, nothing missed.
This free version lets your AI see the video. crv Pro lets it understand it — how it was shot (cut rhythm, camera moves) plus a timestamped timeline of what frames can't show: gestures, expressions, voice pitch shifts, emotion, sound events. One-time founder price.
Most AI tools don't really see a video. Paste a YouTube link into ChatGPT and it
reads the transcript , not the picture. Claude won't take a video file at all.
Even Gemini, which can read video natively, has to send it up to Google and
samples frames at a fixed interval (1 fps by default), so fast cuts slip past.
claude-real-video does it differently, and the processing runs locally : point it at a URL or a
file, and it pulls the frames that actually matter (every scene change, not a
fixed quota), throws away the near-duplicates, transcribes the audio, and hands
you a clean folder any LLM can read. All the processing happens on your own machine — what gets sent anywhere is only the frames/text you choose to paste into an LLM afterwards.
crv " https://www.youtube.com/watch?v=... "
# → crv-out/frames/*.jpg + crv-out/transcript.txt + crv-out/MANIFEST.txt
Then drop the frames + MANIFEST.txt into Claude / ChatGPT / Gemini and ask away.
No terminal needed — run crv-web and a local page opens (Traditional Chinese / Simplified Chinese / English): paste a YouTube or Reels link or a file path, click Analyze, open the result viewer. Runs 100% locally, nothing uploaded.
Want to eyeball what the model will see first? Add --viewer — it writes a local viewer.html (video + keyframe grid + transcript) you can double-click open. No network, no extra installs.
Slow-changing content (animation tutorials, gradual morphs, slow pans): add --adaptive — frames are picked against their rolling neighbourhood instead of a fixed threshold, so a 2-3s squash-and-stretch that never spikes any single frame still gets captured.
Text-heavy content (burned-in captions, lecture slides, screen recordings): add --text-anchors — extra frames are forced at subtitle-cue timestamps (sidecar .srt / .vtt or embedded track), so each spoken segment gets a matching visual even when the scene barely changes. At most one forced frame per second; scene detection is untouched.
Not doing LLM work? It also works as a general-purpose video keyframe extractor —
scene-change detection + dedup, no ML
The roast
Your idea of letting LLMs 'watch' videos locally is interesting, but the market for this is murky at best. You claim speed as your moat (q11=speed), yet you're relying on local processing, which can vary greatly across different consumer hardware. Additionally, while you point out the limitations of current AI tools, you fail to address why users would switch to your solution beyond mere novelty. The execution plausibility is shaky with your solo (q13=solo) and unfunded (q14=no_funding) setup. Your tech might be impressive, but can you really convince users to pay?
Red flags
- q13=solo
- q14=no_funding
- q15=will_pay
Verdict
You've built a smart tool for a niche problem, but convincing a general audience to pay for it will be an uphill battle.
Roast your own startup idea →