2023/10/23 by Xinyu Huang, Huang, Xinyu, Youcai Zhang +14 · 15 citations
Computer Science · #Multimodal Machine Learning Applications #Topic Modeling #Image Retrieval and Classification Techniques
paper · pdf · doi:10.48550/arxiv.2310.15200
In this paper, we introduce the Recognize Anything Plus Model (RAM++), an open-set image tagging model effectively leveraging multi-grained text supervision. Previous approaches (e.g., CLIP) primarily utilize global text supervision paired with images, leading to sub-optimal performance in recognizing multiple individual semantic tags. In contrast, RAM++ seamlessly integrates individual tag supervision with global text supervision, all within a unified alignment framework. This integration not only ensures efficient recognition of predefined tag categories, but also enhances generalization capabilities for diverse open-set categories. Furthermore, RAM++ employs large language models (LLMs) to convert semantically constrained tag supervision into more expansive tag description supervision, thereby enriching the scope of open-set visual description concepts. Comprehensive evaluations on various image recognition benchmarks demonstrate RAM++ exceeds existing state-of-the-art (SOTA) open-set image tagging models on most aspects. Specifically, for predefined commonly used tag categories, RAM++ showcases 10.2 mAP and 15.4 mAP enhancements over CLIP on OpenImages and ImageNet. For open-set categories beyond predefined, RAM++ records improvements of 5.0 mAP and 6.4 mAP over CLIP and RAM respectively on OpenImages. For diverse human-object interaction phrases, RAM++ achieves 7.8 mAP and 4.7 mAP improvements on the HICO benchmark. Code, datasets and pre-trained models are available at \urlhttps://github.com/xinyu1205/recognize-anything.