Self-Supervised Representation Learning via Neighborhood-Relational Encoding

TL;DR

This paper introduces a neighborhood-relational encoding (NRE) method for self-supervised learning that preserves local data structure, improving performance across various vision tasks without heavy annotation.

Contribution

The novel NRE approach captures local neighborhood relations in data, enhancing self-supervised representation learning for multiple vision applications.

Findings

Outperforms or matches state-of-the-art in classification, detection, and segmentation.

Effective in defending against adversarial attacks.

Useful for video anomaly detection.

Abstract

In this paper, we propose a novel self-supervised representation learning by taking advantage of a neighborhood-relational encoding (NRE) among the training data. Conventional unsupervised learning methods only focused on training deep networks to understand the primitive characteristics of the visual data, mainly to be able to reconstruct the data from a latent space. They often neglected the relation among the samples, which can serve as an important metric for self-supervision. Different from the previous work, NRE aims at preserving the local neighborhood structure on the data manifold. Therefore, it is less sensitive to outliers. We integrate our NRE component with an encoder-decoder structure for learning to represent samples considering their local neighborhood information. Such discriminative and unsupervised representation learning scheme is adaptable to different computer vision tasks due to its independence from intense annotation requirements. We evaluate our proposed method for different tasks, including classification, detection, and segmentation based on the learned latent representations. In addition, we adopt the auto-encoding capability of our proposed method for applications like defense against adversarial example attacks and video anomaly detection. Results confirm the performance of our method is better or at least comparable with the state-of-the-art for each specific application, but with a generic and self-supervised approach.