marialymperaiou/visual-genome-embeddings
Visual Genome word embeddings on region descriptions
This project helps image analysts or researchers organize and find images based on their descriptive text. It takes a collection of images, each with natural language descriptions of its regions, and transforms these descriptions into numerical codes. These codes then allow for grouping similar images or retrieving images that match a textual query, even if the descriptions are a bit vague. It's for anyone working with large visual datasets that need semantic organization.
No commits in the last 6 months.
Use this if you need to semantically cluster images or retrieve specific images using either another image's content or a descriptive text phrase.
Not ideal if your images lack detailed text descriptions or if you need to retrieve images based on pixel-level features rather than conceptual similarity.
Stars
4
Forks
—
Language
Jupyter Notebook
License
—
Category
Last pushed
Jun 04, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/marialymperaiou/visual-genome-embeddings"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
dselivanov/text2vec
Fast vectorization, topic modeling, distances and GloVe word embeddings in R.
ibrahimsharaf/doc2vec
:notebook: Long(er) text representation and classification using Doc2Vec embeddings
avidale/compress-fasttext
Tools for shrinking fastText models (in gensim format)
iarroyof/sentence_embedding
A sentence embedding method based on weighted series
awslabs/sagemaker-privacy-for-nlp
A solution that helps apply a privacy preserving mechanism to NLP data, using Amazon SageMaker.