Text-Guided Image Clustering
This work addresses image clustering for computer vision researchers by proposing a novel text-based paradigm, though it is incremental in leveraging existing captioning and VQA models.
The paper tackles image clustering by using generated text from image captioning and VQA models instead of traditional image features, showing that text representations often outperform image features across eight datasets. It also introduces a method to inject domain knowledge via VQA prompting and a counting-based explainability approach that improves cluster descriptions.
Image clustering divides a collection of images into meaningful groups, typically interpreted post-hoc via human-given annotations. Those are usually in the form of text, begging the question of using text as an abstraction for image clustering. Current image clustering methods, however, neglect the use of generated textual descriptions. We, therefore, propose Text-Guided Image Clustering, i.e., generating text using image captioning and visual question-answering (VQA) models and subsequently clustering the generated text. Further, we introduce a novel approach to inject task- or domain knowledge for clustering by prompting VQA models. Across eight diverse image clustering datasets, our results show that the obtained text representations often outperform image features. Additionally, we propose a counting-based cluster explainability method. Our evaluations show that the derived keyword-based explanations describe clusters better than the respective cluster accuracy suggests. Overall, this research challenges traditional approaches and paves the way for a paradigm shift in image clustering, using generated text.