How to find similar images with Vision transformers (ViT) in PyTorch? I show you a simple code implementation where we apply a pre-trained Vision Transformer, which has been fine-tuned on a image dataset (publicly available "beans" dataset) to find similar images given a query image, in a set of thousands of other images. We use Huggingface's models and dataset library for demonstration. Find similar pictures with Vision transformer technology. Official links: https://huggingface.co/blog/image-similarity https://huggingface.co/nateraw/vit-base-beans https://github.com/google-research/vision_transformer Jupyter NB: https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/image_similarity.ipynb #ai #imagesearch #similarity