Generative Adversarial Models for People Attribute Recognition in Surveillance

TL;DR

This paper introduces a deep generative model that enhances surveillance images to improve recognition of people's attributes like gender and clothing, even under poor resolution and occlusion conditions.

Contribution

It presents a novel deep architecture combining generative image enhancement with attribute classification for surveillance footage.

Findings

Effective attribute recognition with up to 80% occlusion

Improved performance on low-resolution surveillance images

Demonstrated robustness against occlusion and resolution issues

Abstract

In this paper we propose a deep architecture for detecting people attributes (e.g. gender, race, clothing ...) in surveillance contexts. Our proposal explicitly deal with poor resolution and occlusion issues that often occur in surveillance footages by enhancing the images by means of Deep Convolutional Generative Adversarial Networks (DCGAN). Experiments show that by combining both our Generative Reconstruction and Deep Attribute Classification Network we can effectively extract attributes even when resolution is poor and in presence of strong occlusions up to 80\% of the whole person figure.