Artificial Dummies for Urban Dataset Augmentation

TL;DR

This paper introduces DummyNet, a novel data generator for urban scene augmentation that creates diverse, rare, and challenging pedestrian images to improve detector performance, especially in difficult conditions like night-time.

Contribution

The paper presents a new network architecture for controlled synthesis of urban scenes with pedestrians, enhancing training data diversity for better detector accuracy.

Findings

DummyNet improves pedestrian detector performance across multiple datasets.

The method enhances detection accuracy in challenging scenarios like night-time conditions.

A 17% reduction in log-average miss rate for night-time detection when trained with DummyNet-augmented data.

Abstract

Existing datasets for training pedestrian detectors in images suffer from limited appearance and pose variation. The most challenging scenarios are rarely included because they are too difficult to capture due to safety reasons, or they are very unlikely to happen. The strict safety requirements in assisted and autonomous driving applications call for an extra high detection accuracy also in these rare situations. Having the ability to generate people images in arbitrary poses, with arbitrary appearances and embedded in different background scenes with varying illumination and weather conditions, is a crucial component for the development and testing of such applications. The contributions of this paper are three-fold. First, we describe an augmentation method for controlled synthesis of urban scenes containing people, thus producing rare or never-seen situations. This is achieved with a data generator (called DummyNet) with disentangled control of the pose, the appearance, and the target background scene. Second, the proposed generator relies on novel network architecture and associated loss that takes into account the segmentation of the foreground person and its composition into the background scene. Finally, we demonstrate that the data generated by our DummyNet improve performance of several existing person detectors across various datasets as well as in challenging situations, such as night-time conditions, where only a limited amount of training data is available. In the setup with only day-time data available, we improve the night-time detector by $17\%$ log-average miss rate over the detector trained with the day-time data only.