ParGANDA: Making Synthetic Pedestrians A Reality For Object Detection

TL;DR

This paper introduces ParGANDA, a method that uses GAN-based image translation to improve synthetic pedestrian data realism for object detection, reducing the domain gap without needing real data labels.

Contribution

The paper presents a novel GAN-based approach for unpaired image translation that maintains annotation accuracy, enhancing synthetic data utility for pedestrian detection tasks.

Findings

Improved detection performance using synthetic data with GAN translation.

Generated images are visually plausible and domain-adapted.

Method eliminates the need for real data labels in training.

Abstract

Object detection is the key technique to a number of Computer Vision applications, but it often requires large amounts of annotated data to achieve decent results. Moreover, for pedestrian detection specifically, the collected data might contain some personally identifiable information (PII), which is highly restricted in many countries. This label intensive and privacy concerning task has recently led to an increasing interest in training the detection models using synthetically generated pedestrian datasets collected with a photo-realistic video game engine. The engine is able to generate unlimited amounts of data with precise and consistent annotations, which gives potential for significant gains in the real-world applications. However, the use of synthetic data for training introduces a synthetic-to-real domain shift aggravating the final performance. To close the gap between the real and synthetic data, we propose to use a Generative Adversarial Network (GAN), which performsparameterized unpaired image-to-image translation to generate more realistic images. The key benefit of using the GAN is its intrinsic preference of low-level changes to geometric ones, which means annotations of a given synthetic image remain accurate even after domain translation is performed thus eliminating the need for labeling real data. We extensively experimented with the proposed method using MOTSynth dataset to train and MOT17 and MOT20 detection datasets to test, with experimental results demonstrating the effectiveness of this method. Our approach not only produces visually plausible samples but also does not require any labels of the real domain thus making it applicable to the variety of downstream tasks.