Home > Research > Publications & Outputs > NOVA

Electronic data

  • NOVA_Kerim_AuthorVersion

    Rights statement: This is the peer reviewed version of the following article: Kerim, A., Aslan, C., Celikcan, U., Erdem, E. and Erdem, A. (2021), NOVA: Rendering Virtual Worlds with Humans for Computer Vision Tasks. Computer Graphics Forum, 40: 258-272. https://doi.org/10.1111/cgf.14271 which has been published in final form at https://onlinelibrary.wiley.com/doi/10.1111/cgf.14271 This article may be used for non-commercial purposes in accordance With Wiley Terms and Conditions for self-archiving.

    Accepted author manuscript, 57.1 MB, PDF document

    Embargo ends: 8/05/22

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

NOVA: Rendering Virtual Worlds with Humans for Computer Vision Tasks

Research output: Contribution to journalJournal articlepeer-review

E-pub ahead of print
  • A. Kerim
  • C. Aslan
  • U. Celikcan
  • E. Erdem
  • A. Erdem
Close
<mark>Journal publication date</mark>8/05/2021
<mark>Journal</mark>Computer Graphics Forum
Issue number6
Volume40
Number of pages15
Pages (from-to)258-272
Publication StatusE-pub ahead of print
Early online date8/05/21
<mark>Original language</mark>English

Abstract

Today, the cutting edge of computer vision research greatly depends on the availability of large datasets, which are critical for effectively training and testing new methods. Manually annotating visual data, however, is not only a labor-intensive process but also prone to errors. In this study, we present NOVA, a versatile framework to create realistic-looking 3D rendered worlds containing procedurally generated humans with rich pixel-level ground truth annotations. NOVA can simulate various environmental factors such as weather conditions or different times of day, and bring an exceptionally diverse set of humans to life, each having a distinct body shape, gender and age. To demonstrate NOVA's capabilities, we generate two synthetic datasets for person tracking. The first one includes 108 sequences, each with different levels of difficulty like tracking in crowded scenes or at nighttime and aims for testing the limits of current state-of-the-art trackers. A second dataset of 97 sequences with normal weather conditions is used to show how our synthetic sequences can be utilized to train and boost the performance of deep-learning based trackers. Our results indicate that the synthetic data generated by NOVA represents a good proxy of the real-world and can be exploited for computer vision tasks.

Bibliographic note

This is the peer reviewed version of the following article: Kerim, A., Aslan, C., Celikcan, U., Erdem, E. and Erdem, A. (2021), NOVA: Rendering Virtual Worlds with Humans for Computer Vision Tasks. Computer Graphics Forum, 40: 258-272. https://doi.org/10.1111/cgf.14271 which has been published in final form at https://onlinelibrary.wiley.com/doi/10.1111/cgf.14271 This article may be used for non-commercial purposes in accordance With Wiley Terms and Conditions for self-archiving.