DisCont: Self-Supervised Visual Attribute Disentanglement using Context Vectors
This addresses the challenge of interpretability and control in image analysis for researchers and practitioners, but it appears incremental as it builds on existing contrastive learning paradigms.
The paper tackles the problem of unsupervised visual attribute disentanglement by proposing DisCont, a self-supervised framework that exploits structural inductive biases in images, and it demonstrates efficacy on four benchmark datasets.
Disentangling the underlying feature attributes within an image with no prior supervision is a challenging task. Models that can disentangle attributes well provide greater interpretability and control. In this paper, we propose a self-supervised framework DisCont to disentangle multiple attributes by exploiting the structural inductive biases within images. Motivated by the recent surge in contrastive learning paradigms, our model bridges the gap between self-supervised contrastive learning algorithms and unsupervised disentanglement. We evaluate the efficacy of our approach, both qualitatively and quantitatively, on four benchmark datasets.