Generation of Synthetic Images for Pedestrian Detection Using a Sequence of GANs

TL;DR

This paper introduces a novel pipeline using a sequence of three GANs to generate synthetic images for pedestrian detection, significantly improving detection performance despite less visually appealing images.

Contribution

It presents a new method combining multiple GANs in sequence to augment datasets for pedestrian detection, a novel approach at the time.

Findings

Detection benchmark surpasses baseline performance

Synthetic images improve pedestrian detection accuracy

Pipeline demonstrates potential despite visual quality issues

Abstract

Creating annotated datasets demands a substantial amount of manual effort. In this proof-of-concept work, we address this issue by proposing a novel image generation pipeline. The pipeline consists of three distinct generative adversarial networks (previously published), combined in a novel way to augment a dataset for pedestrian detection. Despite the fact that the generated images are not always visually pleasant to the human eye, our detection benchmark reveals that the results substantially surpass the baseline. The presented proof-of-concept work was done in 2020 and is now published as a technical report after a three years retention period.