VELDA: Relating an Image Tweet’s Text and Images


  • Tao Chen National University of Singapore
  • Hany SalahEldeen Old Dominion University
  • Xiangnan He National University of Singapore
  • Min-Yen Kan National University of Singapore
  • Dongyuan Lu National University of Singapore



image tweets, microblog, image and text, topic model


Image tweets are becoming a prevalent form of socialmedia, but little is known about their content — textualand visual — and the relationship between the two mediums.Our analysis of image tweets shows that while visualelements certainly play a large role in image-text relationships, other factors such as emotional elements, also factor into the relationship. We develop Visual-Emotional LDA (VELDA), a novel topic model to capturethe image-text correlation from multiple perspectives (namely, visual and emotional). Experiments on real-world image tweets in both Englishand Chinese and other user generated content, show that VELDA significantly outperforms existingmethods on cross-modality image retrieval. Even in other domains where emotion does not factor in imagechoice directly, our VELDA model demonstrates good generalization ability, achieving higher fidelity modeling of such multimedia documents.




How to Cite

Chen, T., SalahEldeen, H., He, X., Kan, M.-Y., & Lu, D. (2015). VELDA: Relating an Image Tweet’s Text and Images. Proceedings of the AAAI Conference on Artificial Intelligence, 29(1).