DisCont: Self-Supervised Visual Attribute Disentanglement using Context Vectors

2020/06/10 by Sarthak Bhagat, Vishaal Udandarao, Bhagat, Sarthak +3
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #Digital Media Forensic Detection #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #cs.CV

paper · pdf · doi:10.48550/arxiv.2006.05895

Published at the 37th International Conference on Machine Learning (ICML 2020) Workshop on ML Interpretability for Scientific Discovery

openalex publication_date 2020/06/10 · arxiv created 2020/06/29 · arxiv updated 2020/07/01 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28

Abstract

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.

Related

• Self-Supervised Disentanglement by Leveraging Structure in Data Augmentations
• Understand and Improve Contrastive Learning Methods for Visual Representation: A Review
• A Contrastive Objective for Learning Disentangled Representations
• An Image is Worth More Than a Thousand Words: Towards Disentanglement in the Wild
• Fine-Grained Representation Learning via Multi-Level Contrastive Learning without Class Priors