VALERIE22 -- A photorealistic, richly metadata annotated dataset of urban environments

TL;DR

The paper introduces VALERIE22, a photorealistic synthetic dataset with extensive metadata annotations for urban environments, aimed at improving DNN validation in pedestrian detection for autonomous driving.

Contribution

It presents a novel, richly annotated synthetic dataset generated through a procedural pipeline, enabling detailed analysis of factors influencing DNN perception performance.

Findings

VALERIE22 outperforms several existing datasets in DNN evaluation metrics.

Rich metadata allows detailed analysis of scene and semantic features.

The dataset supports diverse testing scenarios for urban perception models.

Abstract

The VALERIE tool pipeline is a synthetic data generator developed with the goal to contribute to the understanding of domain-specific factors that influence perception performance of DNNs (deep neural networks). This work was carried out under the German research project KI Absicherung in order to develop a methodology for the validation of DNNs in the context of pedestrian detection in urban environments for automated driving. The VALERIE22 dataset was generated with the VALERIE procedural tools pipeline providing a photorealistic sensor simulation rendered from automatically synthesized scenes. The dataset provides a uniquely rich set of metadata, allowing extraction of specific scene and semantic features (like pixel-accurate occlusion rates, positions in the scene and distance + angle to the camera). This enables a multitude of possible tests on the data and we hope to stimulate research on understanding performance of DNNs. Based on performance metric a comparison with several other publicly available datasets is provided, demonstrating that VALERIE22 is one of best performing synthetic datasets currently available in the open domain.