Show HN: Semantic search engine for Studio Ghibli movie
4 days ago (ghibli-search.anini.workers.dev)
Hi HN! I built Ghibli Search, a semantic search engine for Studio Ghibli movie scenes (e.g. Spirited Away, My Neighbor Totoro, Howl's Moving Castle, etc.).
Describe a dreamscape like "flying through clouds at sunset" or upload an image, and it finds visually similar scenes from the films.
Live demo: https://ghibli-search.anini.workers.dev/
Full Cloudflare stack: Workers, AI Search, R2, Workers AI
Open source: https://github.com/aninibread/ghibli-search
Would love feedback on the search quality and any ideas for improvements!
It's interesting that searches for general concepts (Countryside wildflowers, airplane scenes, train scenes) seem to work pretty well, while searches for specific characters (Ponyo, Yubaba, Porco Rosso) return negligible results.
That's great feedback. The way it works right now is that I use Cloudflare AI Search to convert my image to text and embed the images. https://developers.cloudflare.com/ai-search/concepts/how-ai-...
Most of the descriptions generated for the images are likely generic and doesn't contain reference to the actual characters and names. I might explore incorporating some sort of more movie specific description in the metadata of the image to improve queries for specific characters.
“Malevolent spirits waiting at the bottom of a pit”
Didn’t return what I expected: the scene from spirited away when chihiro and haku fall through the trap door.
Relevance can be improved with multiple ranking steps.
Curious why specifically add multiple ranking steps to improve relevance? How would each ranking step differ ? Would you use different ranking models?
curious: can anyone use Ghibli's movie scenes on a random website just like that?
The site doesn't stream the movies, references still frames and the original works, and links directly back to the official site - there's no exploitation or arbitrage taking anything away from the studio.
Kelly v. Arriba Soft Corp. (2003) and Perfect 10 v. Amazon (2007) are precedents for image search engines displaying thumbnails - they were found to be fair use. The function is transformative, the site is for a completely different use case than watching media, and doesn't harm the market.
If they've purchased the movies legitimately, and have the receipts, they have an incredibly strong fair use case. Because it's beneficial to Studio Ghibli, I'd say they are best served by allowing it and not trying to exploit DMCA mechanisms to get them taken down.
This is one of those areas where copyright holders can be assholes and abuse the system for petty wins, but the big tech companies have fought and won explicit precedent demonstrating the legitimacy of fair use cases for tools exactly like this.
Awesome tool!
> If they've purchased the movies legitimately, and have the receipts, they have an incredibly strong fair use case.
While I'd also argue that this could be covered under a fair use defense, I thought it worth pointing out that buying a copy of a work and having receipts would have no bearing on the right to distribute copies of that work to others.
Obviously, if someone pirated these movies they could get in trouble for that as well, but that'd be an entirely different matter from the use of copyrighted images on their website.
1 reply →
@dang this domain should probably be expanded from showing just workers.dev
[dead]