-
CLIP
CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
-
fastdup
fastdup is a powerful free tool designed to rapidly extract valuable insights from your image & video datasets. Assisting you to increase your dataset images & labels quality and reduce your data operations costs at an unparalleled scale.
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
One way could be using the CLIP model from OpenAI (quite easy) to get an image vector for each, and then using simple cosine similarity to compare the vectors.
I came across fastdup recently https://github.com/visual-layer/fastdup
Related posts
-
Visualize your dataset using DINOv2 embedding
-
Visualize your dataset using DINOv2 embedding
-
[R][P] How to extract feature vectors of large datasets using DINOv2 on CPU
-
Computer Vision pre-trained model for finding how similar two photos of a room are
-
Find image duplicates and outliers – A free, scalable, efficient tool