Learning Transferable Visual Models From Natural Language Supervision

by   Alec Radford, et al.

State-of-the-art computer vision systems are trained to predict a fixed set of predetermined object categories. This restricted form of supervision limits their generality and usability since additional labeled data is needed to specify any other visual concept. Learning directly from raw text about images is a promising alternative which leverages a much broader source of supervision. We demonstrate that the simple pre-training task of predicting which caption goes with which image is an efficient and scalable way to learn SOTA image representations from scratch on a dataset of 400 million (image, text) pairs collected from the internet. After pre-training, natural language is used to reference learned visual concepts (or describe new ones) enabling zero-shot transfer of the model to downstream tasks. We study the performance of this approach by benchmarking on over 30 different existing computer vision datasets, spanning tasks such as OCR, action recognition in videos, geo-localization, and many types of fine-grained object classification. The model transfers non-trivially to most tasks and is often competitive with a fully supervised baseline without the need for any dataset specific training. For instance, we match the accuracy of the original ResNet-50 on ImageNet zero-shot without needing to use any of the 1.28 million training examples it was trained on. We release our code and pre-trained model weights at https://github.com/OpenAI/CLIP.



There are no comments yet.


page 13

page 15

page 24

page 42


K-LITE: Learning Transferable Visual Models with External Knowledge

Recent state-of-the-art computer vision systems are trained from natural...

CLIP-Art: Contrastive Pre-Training for Fine-Grained Art Classification

Existing computer vision research in artwork struggles with artwork's fi...

Learning to Prompt for Vision-Language Models

Vision-language pre-training has recently emerged as a promising alterna...

Evaluating CLIP: Towards Characterization of Broader Capabilities and Downstream Implications

Recently, there have been breakthroughs in computer vision ("CV") models...

An implementation of the "Guess who?" game using CLIP

CLIP (Contrastive Language-Image Pretraining) is an efficient method for...

Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm

Recently, large-scale Contrastive Language-Image Pre-training (CLIP) has...

PyramidCLIP: Hierarchical Feature Alignment for Vision-language Model Pretraining

Large-scale vision-language pre-training has achieved promising results ...

Code Repositories


Generate vector graphics from a textual caption

view repo


CLIP implementation for Russian language

view repo


Repository for experiments with several approaches to fine-tune model, pretrained on the CLIP: https://openai.com/blog/clip/

view repo


OpenAI clip implementation in PaddlePaddle

view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.