Improving Image Clustering using Sparse Text and the Wisdom of the Crowds
This work addresses image clustering for domains with sparse text data, but it appears incremental as it combines existing techniques like feature fusion and matrix factorization.
The paper tackled the problem of image clustering by fusing image and sparse text features using a common dictionary based on crowd wisdom, and applied non-negative matrix factorization for topic modeling to cluster documents, but no concrete results or numbers were provided.
We propose a method to improve image clustering using sparse text and the wisdom of the crowds. In particular, we present a method to fuse two different kinds of document features, image and text features, and use a common dictionary or "wisdom of the crowds" as the connection between the two different kinds of documents. With the proposed fusion matrix, we use topic modeling via non-negative matrix factorization to cluster documents.