Semantically Selective Augmentation for Deep Compact Person Re-Identification

TL;DR

This paper introduces a deep person re-identification method that uses semantically controlled data augmentation with GANs and network compression to achieve high accuracy with lightweight models, outperforming existing methods.

Contribution

It proposes a novel semantically selective data augmentation technique using a constrained DCGAN and combines it with clustering-based network compression for improved re-identification.

Findings

Outperforms state-of-the-art on LIMA dataset

Effective semantic control improves data augmentation quality

Network compression yields faster inference without accuracy loss

Abstract

We present a deep person re-identification approach that combines semantically selective, deep data augmentation with clustering-based network compression to generate high performance, light and fast inference networks. In particular, we propose to augment limited training data via sampling from a deep convolutional generative adversarial network (DCGAN), whose discriminator is constrained by a semantic classifier to explicitly control the domain specificity of the generation process. Thereby, we encode information in the classifier network which can be utilized to steer adversarial synthesis, and which fuels our CondenseNet ID-network training. We provide a quantitative and qualitative analysis of the approach and its variants on a number of datasets, obtaining results that outperform the state-of-the-art on the LIMA dataset for long-term monitoring in indoor living spaces.