Recognize Anything: A Strong Image Tagging Model

TL;DR

The Recognize Anything Model (RAM) is a new foundation model for image tagging that achieves high zero-shot recognition accuracy across common categories by leveraging large-scale image-text data and a multi-step training process.

Contribution

RAM introduces a novel training paradigm for image tagging using automatic annotation and data cleaning, significantly improving zero-shot performance over existing models.

Findings

RAM outperforms CLIP and BLIP on multiple benchmarks.

RAM surpasses fully supervised methods in accuracy.

RAM achieves performance comparable to Google tagging API.

Abstract

We present the Recognize Anything Model (RAM): a strong foundation model for image tagging. RAM makes a substantial step for large models in computer vision, demonstrating the zero-shot ability to recognize any common category with high accuracy. RAM introduces a new paradigm for image tagging, leveraging large-scale image-text pairs for training instead of manual annotations. The development of RAM comprises four key steps. Firstly, annotation-free image tags are obtained at scale through automatic text semantic parsing. Subsequently, a preliminary model is trained for automatic annotation by unifying the caption and tagging tasks, supervised by the original texts and parsed tags, respectively. Thirdly, a data engine is employed to generate additional annotations and clean incorrect ones. Lastly, the model is retrained with the processed data and fine-tuned using a smaller but higher-quality dataset. We evaluate the tagging capabilities of RAM on numerous benchmarks and observe impressive zero-shot performance, significantly outperforming CLIP and BLIP. Remarkably, RAM even surpasses the fully supervised manners and exhibits competitive performance with the Google tagging API. We are releasing the RAM at \url{https://recognize-anything.github.io/} to foster the advancements of large models in computer vision.