Unsupervised Correlation Analysis
This addresses a fundamental limitation in computer vision by enabling domain linking without correspondences, though it is incremental as it builds on CCA methods.
The paper tackled the problem of linking data from different domains without requiring prior correspondences, introducing Unsupervised Correlation Analysis (UCA) that replaces correlation maximization with reconstruction and domain confusion terms, and showed it can link remote domains like text and images while approaching supervised performance on benchmarks.
Linking between two data sources is a basic building block in numerous computer vision problems. In this paper, we set to answer a fundamental cognitive question: are prior correspondences necessary for linking between different domains? One of the most popular methods for linking between domains is Canonical Correlation Analysis (CCA). All current CCA algorithms require correspondences between the views. We introduce a new method Unsupervised Correlation Analysis (UCA), which requires no prior correspondences between the two domains. The correlation maximization term in CCA is replaced by a combination of a reconstruction term (similar to autoencoders), full cycle loss, orthogonality and multiple domain confusion terms. Due to lack of supervision, the optimization leads to multiple alternative solutions with similar scores and we therefore introduce a consensus-based mechanism that is often able to recover the desired solution. Remarkably, this suffices in order to link remote domains such as text and images. We also present results on well accepted CCA benchmarks, showing that performance far exceeds other unsupervised baselines, and approaches supervised performance in some cases.