Learning Inward Scaled Hypersphere Embedding: Exploring Projections in Higher Dimensions

TL;DR

This paper introduces inward scaled hypersphere embedding, a novel approach that enhances feature discrimination in high-dimensional spaces, improving classification and retrieval performance.

Contribution

It proposes a new inward scaling technique for hypersphere embedding and a simpler CNN architecture, advancing discriminative feature analysis in high-dimensional spaces.

Findings

Achieved results comparable to state-of-the-art methods in classification.

Improved retrieval accuracy with the proposed embedding technique.

Demonstrated effectiveness of inward scaling in high-dimensional feature spaces.

Abstract

Majority of the current dimensionality reduction or retrieval techniques rely on embedding the learned feature representations onto a computable metric space. Once the learned features are mapped, a distance metric aids the bridging of gaps between similar instances. Since the scaled projection is not exploited in these methods, discriminative embedding onto a hyperspace becomes a challenge. In this paper, we propose to inwardly scale feature representations in proportional to projecting them onto a hypersphere manifold for discriminative analysis. We further propose a novel, yet simpler, convolutional neural network based architecture and extensively evaluate the proposed methodology in the context of classification and retrieval tasks obtaining results comparable to state-of-the-art techniques.