Multilinear subspace learning for person re-identification based fusion of high order tensor features

TL;DR

This paper introduces a multilinear subspace learning approach that fuses high-order tensor features from CNN and LOMO for improved person re-identification, demonstrating superior performance on multiple datasets.

Contribution

It proposes a novel tensor fusion scheme and multilinear subspace learning method (TXQDA) for combining diverse features in person re-identification tasks.

Findings

Outperforms recent state-of-the-art methods on VIPeR, GRID, and PRID450S datasets.

Effectively fuses features from CNN and LOMO using tensor methods.

Reduces high dimensionality of tensor data while maintaining discriminative power.

Abstract

Video surveillance image analysis and processing is a challenging field in computer vision, with one of its most difficult tasks being Person Re-Identification (PRe-ID). PRe-ID aims to identify and track target individuals who have already been detected in a network of cameras, using a robust description of their pedestrian images. The success of recent research in person PRe-ID is largely due to effective feature extraction and representation, as well as the powerful learning of these features to reliably discriminate between pedestrian images. To this end, two powerful features, Convolutional Neural Networks (CNN) and Local Maximal Occurrence (LOMO), are modeled on multidimensional data using the proposed method, High-Dimensional Feature Fusion (HDFF). Specifically, a new tensor fusion scheme is introduced to leverage and combine these two types of features in a single tensor, even though their dimensions are not identical. To enhance the system's accuracy, we employ Tensor Cross-View Quadratic Analysis (TXQDA) for multilinear subspace learning, followed by cosine similarity for matching. TXQDA efficiently facilitates learning while reducing the high dimensionality inherent in high-order tensor data. The effectiveness of our approach is verified through experiments on three widely-used PRe-ID datasets: VIPeR, GRID, and PRID450S. Extensive experiments demonstrate that our approach outperforms recent state-of-the-art methods.