The pursuit of beauty: Converting image labels to meaningful vectors

TL;DR

This paper introduces OLR, a method that converts image labels into meaningful, disentangled low-dimensional vectors capturing data semantics, aiding image understanding and analysis.

Contribution

The paper presents a novel approach, OLR, for transforming image labels into informative, disentangled representations that reflect data concepts and interrelations.

Findings

OLR captures significant data semantics.

OLR discovers data interrelations.

Representations are disentangled and low-dimensional.

Abstract

A challenge of the computer vision community is to understand the semantics of an image, in order to allow image reconstruction based on existing high-level features or to better analyze (semi-)labelled datasets. Towards addressing this challenge, this paper introduces a method, called Occlusion-based Latent Representations (OLR), for converting image labels to meaningful representations that capture a significant amount of data semantics. Besides being informational rich, these representations compose a disentangled low-dimensional latent space where each image label is encoded into a separate vector. We evaluate the quality of these representations in a series of experiments whose results suggest that the proposed model can capture data concepts and discover data interrelations.