LCrowdV: Generating labeled videos for pedestrian detectors training and crowd behavior learning

NEUROCOMPUTING

卷:337

Article

Cheung, Ernest¹  Wong, Anson¹  Bera, Aniket¹  Wang, Xiaogang²  Manocha, Dinesh³ 

[1] Univ North Carolina Chapel Hill, Chapel Hill, NC 27515 USA

[2] Chinese Univ Hong Kong, Hong Kong, Peoples R China

[3] Univ Maryland Coll Pk, College Pk, MD USA

关键词: Crowd analysis;    Pedestrian detection;    Crowd behaviors;    Crowd datasets;    Crowd simulation;    Crowd rendering;   

DOI  :  10.1016/j.neucom.2018.08.085

来源: Elsevier

【 摘 要 】

We present a novel procedural framework to generate an arbitrary number of labeled crowd videos (LCrowdV). The resulting crowd video datasets are used to design accurate algorithms or training models for crowded scene understanding. Our overall approach is composed of two components: a procedural simulation framework to generate crowd movements and behaviors, and a procedural rendering framework to generate different videos or images. Each video or image is automatically labeled based on the environment, number of pedestrians, density, behavior (personality), flow, lighting conditions, viewpoint, type of noise, etc. Furthermore, we can increase the realism by combining synthetically-generated behaviors with real-world background videos. We demonstrate the benefits of LCrowdV over prior lableled crowd datasets by augmenting a real dataset with it and improving the accuracy in pedestrian detection and crowd classification. Furthermore, we evaluate the impact of removing the variety in different LCrowdV parameters to show the importance of the diversity of data generated from our framework. LCrowdV has been made available as an online resource. (c) 2018 Published by Elsevier B.V.

【 授权许可】

Free