TechPrufer
Get started
Replicate
Replicate logoReplicate · Feature trail

Image Collection and Retrieval

1 post · Aug 2022

What this trail covers

Replicate has introduced tools and techniques for automating image collection and retrieval, leveraging CLIP embeddings and k-NN search over large datasets like LAION-5B. This enables users to curate custom datasets for fine-tuning models, steer text-to-image generation with specific aesthetics, and find images similar to a given input image. The `clip-retrieval` Python package is highlighted as a key tool for this capability, with integrations into Replicate's platform for running text-to-image generation and for finding images similar to a given input image.

All Replicate trails
  1. Automating image collectionThis post details how to use the `clip-retrieval` Python package to query the LAION-5B dataset using text prompts or image embeddings. It provides installation instructions, code examples for querying with text and images, and demonstrates how to use the retrieved images as initial images for text-to-image models on Replicate. It also explains the underlying technical concepts of CLIP embeddings and k-NN search.