Information Fusion: Scaling Subspace-Driven Approaches
This work addresses multimodal data clustering, which is important for applications like computer vision and sensor fusion, but it appears incremental as it builds on existing subspace clustering methods.
The paper tackles multimodal data analysis by developing a deep subspace clustering method called DRoGSuRe, which exploits group subspace distributions using CNNs and optimized integration, showing it is competitive with and more robust to noise than the state-of-the-art DMSC approach on various datasets.
In this work, we seek to exploit the deep structure of multi-modal data to robustly exploit the group subspace distribution of the information using the Convolutional Neural Network (CNN) formalism. Upon unfolding the set of subspaces constituting each data modality, and learning their corresponding encoders, an optimized integration of the generated inherent information is carried out to yield a characterization of various classes. Referred to as deep Multimodal Robust Group Subspace Clustering (DRoGSuRe), this approach is compared against the independently developed state-of-the-art approach named Deep Multimodal Subspace Clustering (DMSC). Experiments on different multimodal datasets show that our approach is competitive and more robust in the presence of noise.