Text-Image Retrieval using CLIP

[Paper]

To use the text-image retrieval, run file: Zero-shot text-image retrieval using CLIP.ipynb

Environment setup

$ conda install pytorch pytorch=1.7.1 torchvision cudatoolkit=11.1
$ pip install ftfy regex tqdm
$ pip install git+https://github.com/openai/CLIP.git
$ conda install matplotlib
$ pip install numpy
$ pip install scikit-image

Example result

User given text, retrive image from CIFAR100 dataset

User given text, retrie image from sci-kit images