img2dataset
clip-retrieval
img2dataset | clip-retrieval | |
---|---|---|
13 | 11 | |
3,264 | 2,139 | |
- | - | |
7.1 | 7.7 | |
3 days ago | 18 days ago | |
Python | Jupyter Notebook | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
img2dataset
-
OpenAI sued for web scraping from millions of internet users in order to train ChatGPT
Lmao, no it doesn't. As we can see, their downloader uses very obscure "no ai" headers (which can be disabled, so its useless). They only claim it respects "robots.txt" because the google crawler respects it, if a site changes their robots.txt rules they don't remove it from their dataset, that is not "respecting". https://github.com/rom1504/img2dataset
-
Who kept the bots out? Stopping content being harvested by AI
The particular tool mentioned in the Vice article is Img2dataset, and right now, it doesn't pay attention to the robots.txt file, the normal mechanism you can use to dissuade well behaved bots from indexing your content. However, it does respect a new HTTP header directive, X-Robots-Tag: noai (and also noindex, though that's an existing and already well-known part of the robots.txt standard).
- AI used photographer’s photos for training, then slapped him with an invoice
-
An AI Scraping Tool Is Overwhelming Websites with Traffic
The established norm is that scrapers have to download robots.txt and support the standard robots.txt features, notably including `Crawl-Delay` which sets a rate limit. This is the established standard by which websites tell scrapers what the rules are for scraping them.
This tool is scraping sites, it has webmasters reporting actual disruption, it doesn't have robots.txt support. When people complained (eg in https://github.com/rom1504/img2dataset/issues/48), the author's stance was basically "PRs welcome". It looks like a third party recently contributed a PR to make it respect robots.txt (https://github.com/rom1504/img2dataset/pull/302), albeit without `Crawl-Delay` support, which is not merged yet.
I have seen the same thing with other recent AI tools (eg https://github.com/m1guelpf/browser-agent/issues/2) and I think it's important to defend the robots.txt convention and nip this in the bud. If a bot doesn't make a reasonable effort to respect robots.txt and it causes disruption, it's a denial-of-service attack and should be treated as such. No excuses.
-
Please make this tool “opt-in” by default
//First sentence unchanged
Websites can pass the http headers `X-Robots-Tag: noai`, `X-Robots-Tag: noindex` , `X-Robots-Tag: noimageai` and `X-Robots-Tag: noimageindex` By default img2dataset will ignore images with such headers.
//Then pull up the policy link first
To understand why image creators and artists may choose to specify such headers for their images, and choose to actively not consent to their images being collected, see [AI use impact](https://github.com/rom1504/img2dataset#ai-use-impact).
AI training would not be possible without the contribution of artists, and it is the recommendation of this tool's authors that you should respect their communicated wishes. However, if you have a legitimate reason to bypass these headers, to disable this behavior and download all images, you may pass --disallowed_header_directives '[]'
```
- Img2dataset: Turns large sets of image URLs to an image dataset
-
Stable Attribution
it's not remembering pixels. for it to do that, it would have to have the pixels stored somewhere. It does not.
The laion 5b dataset is in the neighborhood of 220TB. (1) That is how much storage space you need to remember the pixels.
The stable diffusion 1.5 checkpoint is 7gb. (2)
1 https://github.com/rom1504/img2dataset/blob/main/dataset_exa...
2 https://huggingface.co/runwayml/stable-diffusion-v1-5/tree/m...
-
Stability AI plans to let artists opt out of Stable Diffusion 3 image training
AI Bots do not respect these flags anyways
- Rule
-
A Stable Diffusion prompt changes its output for the style of 1500 artists
You can download it here:
https://github.com/rom1504/img2dataset/blob/main/dataset_exa...
You probably would want to stop after getting the metadata, unless you have 240TB available for the images :)
More details and links to dataset explorers here: https://laion.ai/blog/laion-5b/
clip-retrieval
- FLaNK AI for 11 March 2024
-
[D] data for handwriting recognition
The tool clip-retreival lets you filter those 400 million images to whatever subsets you're interested in --- for example, 10,000 images of (mostly) handwriting.
- Stable Attribution
-
Same.energy: Image Search by Similarity
Hehe, well you know, PR welcome, the front end is 500 lines https://github.com/rom1504/clip-retrieval/blob/main/front/sr...
Other people have done a few alternate front ends already
This one is meant to be functional, but could sure be made prettier
-
Is there a way to use clip or blip to search a massive collection of images for specific things within the picture?
This might work: https://github.com/rom1504/clip-retrieval .
-
Ai art
HaveIBeenTrained uses clip retrieval to search the Laion-5B and Laion-400M image datasets. These are currently the largest public text-to-image datsets, and they are used to train models like Stable Diffusion, Imagen, among many others.
- Image Similarity Score using transfer learning
-
Exploring 12M of the 2.3B Images Used to Train Stable Diffusion
Done https://github.com/rom1504/clip-retrieval/commit/53e3383f58b...
Using clip for searching is better than direct text indexing for a variety of reasons but here for example because it matches better what stable diffusion sees
-
Semantic and Similarity Image Search Engine
Based on OpenAI's CLIP and the clip-retrieval library (https://github.com/rom1504/clip-retrieval), I've built an end-to-end demo for a semantic and similarity image search engine. It's incredibly powerful for finding similar images amongst large image datasets, or just submitting text/natural language queries and finding the most relevant images in your dataset. Really useful tool for introspection into large datasets before annotation or ML work begins. This could potentially be used to filter or downsize your datasets by several orders of magnitude and make annotation and ML work easier and less costly.
Checkout the demo here:
http://ec2-52-39-251-116.us-west-2.compute.amazonaws.com/
And you can checkout our website or email me for updates and email list, etc.:
https://machineperception.co
-
What every software engineer should know about search
Assuming you have an NVIDIA GPU, you can build a semantic search engine by indexing CLIP embeds (image or text).
https://github.com/rom1504/clip-retrieval
What are some alternatives?
chitra - A multi-functional library for full-stack Deep Learning. Simplifies Model Building, API development, and Model Deployment.
Typesense - Open Source alternative to Algolia + Pinecone and an Easier-to-Use alternative to ElasticSearch ⚡ 🔍 ✨ Fast, typo tolerant, in-memory fuzzy Search Engine for building delightful search experiences
google-images-download - Python Script to download hundreds of images from 'Google Images'. It is a ready-to-run code!
MoTIS - [NAACL 2022]Mobile Text-to-Image search powered by multimodal semantic representation models(e.g., OpenAI's CLIP)
dalle-mini - DALL·E Mini - Generate images from a text prompt
laion-aesthetic-datasette - Use Datasette to explore LAION improved_aesthetics_6plus training data used by Stable DIffusion
docarray - Represent, send, store and search multimodal data
open_clip - An open source implementation of CLIP.
DeepViewAgg - [CVPR'22 Best Paper Finalist] Official PyTorch implementation of the method presented in "Learning Multi-View Aggregation In the Wild for Large-Scale 3D Semantic Segmentation"
clip-italian - CLIP (Contrastive Language–Image Pre-training) for Italian
stable-diffusion - A latent text-to-image diffusion model
Queryable - Run OpenAI's CLIP model on iOS to search photos.