[HN Gopher] Show HN: I made a Pinterest clone using SigLIP image...
___________________________________________________________________
Show HN: I made a Pinterest clone using SigLIP image embeddings
Click an image to get similar images. I crawled Tumblr and used
SigLIP to get vector embeddings for many images. When you click an
image, it finds the most similar vector embeddings in the database,
and returns the corresponding images.
Author : verse
Score : 77 points
Date : 2024-02-16 02:52 UTC (20 hours ago)
(HTM) web link (mood-amber.vercel.app)
(TXT) w3m dump (mood-amber.vercel.app)
| ijhuygft776 wrote:
| nice, we always need more clones and improvements.... hope you
| get traction.
|
| I never click Pinterest links because the experience is too bad.
| karolist wrote:
| I use unpinterested extension in Chrome to remove pinterest
| from search results, I was annoyed so much at some point. Maybe
| they're SEO spam is more under control now, not sure.
| Yenrabbit wrote:
| Neat! How many images are in the dataset out of curiosity?
| squam wrote:
| Cool project! Thanks for sharing
| Tiberium wrote:
| Is there a repo, especially for training? I'd like to see how
| SigLIP performs on a dataset of only anime images.
| jarebear6expepj wrote:
| The the vision training models are available here:
| https://github.com/google-research/big_vision/tree/main which I
| am assuming, based on the research paper is what was used for
| the project.
| GamerAlias wrote:
| Good stuff! Do you have any intuitive sense of whether SigLIP is
| particularly stronger than CLIP here? Also vector DB over Faiss
| index?
| verse wrote:
| I haven't done much testing or anything, but it seems to me
| that siglip "understands" what it's looking at more than CLIP
|
| also no, I just put everything on Supabase and added pgvector.
| super easy:
|
| https://supabase.com/docs/guides/database/extensions/pgvecto...
| ReD_CoDE wrote:
| qdrant doesn't support vector DB over Faiss index?
|
| Also, pgvector or qdrant? which is better?
| wucaworld wrote:
| Very cool! How did you get the collage layout? I noticed images
| in each column don't have the same size. I assume images get
| Centre cropped?
| jkcxn wrote:
| It's called a masonry grid. Images retain their aspect ratio so
| they don't need to be cropped. You can kind of simulate it with
| css but there's proposals to add a proper masonry layout to css
| verse wrote:
| yeah. I actually wrote the logic for the layout myself
| (wasn't really happy with the available libraries). may open
| source this if people are interested!
| ReD_CoDE wrote:
| Can you share your GH to follow updates? Also, take a look
| at this, they have a layout too https://github.com/lit/lit/
| tree/main/packages/labs/virtualiz...
| verse wrote:
| will post on twitter:
|
| https://x.com/verse_
| lulzx wrote:
| Also, check https://same.energy/
| omeze wrote:
| Cool! I haven't tried SigLIP out yet but it seems to be the new
| hotness over CLIP... I just dont have a good project idea yet
| yorwba wrote:
| Sometimes there are duplicate results, e.g. https://mood-
| amber.vercel.app/images/0b733fc2-7093-4443-8872... has two copies
| of https://mood-
| amber.vercel.app/images/f920a599-bbd7-4805-3317... right next to
| each other. (The link UUID is the same, so I assume this is an
| issue with the search algorithm, not simply duplicate data that
| got scraped.)
| verse wrote:
| ah! thank you for pointing this out. will fix
| convolvatron wrote:
| how far we've come since https://www.karlsims.com/genetic-
| images.html
|
| quite a bit, but surprisingly not
| gammalost wrote:
| There are some interesting images there. Why are you not
| including the source of the images?
___________________________________________________________________
(page generated 2024-02-16 23:00 UTC)