Scalable Image Retrieval with Multimodal Fusion

As the number of images grows rapidly on the Internet, the scalability of image retrieval systems becomes a significant issue. In this paper, we propose two distributed clustering algorithms to scale up the bag-ofvisual-words model on millions of images and billions of visual features by leveraging distributed systems. We also introduce a multimodal fusion model to utilize textual data to improve the quality of image retrieval. Our experiments on multimodal datasets demonstrated our fusion approach can achieve high retrieval quality compared to image-only retrieval and text-only retrieval.