TY - GEN
T1 - NYU-VPR: Long-Term Visual Place Recognition Benchmark with View Direction and Data Anonymization Influences
AU - Sheng, Diwei
AU - Chai, Yuxiang
AU - Li, Xinru
AU - Cheng, Hsiling
AU - Lin, Jianzhe
AU - Silva, Claudio
AU - Rizzo, John-Ross
N1 - Funding Information:
We would like to thank Carmera for providing the raw NYC image data set that we used for creating the NYU-VPR. This research is funded by the Connected Cities for Smart Mobility towards Accessible and Resilient Transportation (C2SMART), a Tier 1 University Center awarded by U.S. Department of Transportation under contract 69A3351747124.
Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Visual place recognition (VPR) is critical in not only localization and mapping for autonomous driving vehicles, but also assistive navigation for the visually impaired population. To enable a long-term VPR system on a large scale, several challenges need to be addressed. First, different applications could require different image view directions, such as front views for self-driving cars while side views for the low vision people. Second, VPR in metropolitan scenes can often cause privacy concerns due to the imaging of pedestrian and vehicle identity information, calling for the need for data anonymization before VPR queries and database construction. Both factors could lead to VPR performance variations that are not well understood yet. To study their influences, we present the NYU-VPR dataset that contains more than 200,000 images over a 2km×2km area near the New York University campus, taken within the whole year of 2016. We present benchmark results on several popular VPR algorithms showing that side views are significantly more challenging for current VPR methods while the influence of data anonymization is almost negligible, together with our hypothetical explanations and in-depth analysis.
AB - Visual place recognition (VPR) is critical in not only localization and mapping for autonomous driving vehicles, but also assistive navigation for the visually impaired population. To enable a long-term VPR system on a large scale, several challenges need to be addressed. First, different applications could require different image view directions, such as front views for self-driving cars while side views for the low vision people. Second, VPR in metropolitan scenes can often cause privacy concerns due to the imaging of pedestrian and vehicle identity information, calling for the need for data anonymization before VPR queries and database construction. Both factors could lead to VPR performance variations that are not well understood yet. To study their influences, we present the NYU-VPR dataset that contains more than 200,000 images over a 2km×2km area near the New York University campus, taken within the whole year of 2016. We present benchmark results on several popular VPR algorithms showing that side views are significantly more challenging for current VPR methods while the influence of data anonymization is almost negligible, together with our hypothetical explanations and in-depth analysis.
UR - http://www.scopus.com/inward/record.url?scp=85124352623&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85124352623&partnerID=8YFLogxK
U2 - 10.1109/IROS51168.2021.9636640
DO - 10.1109/IROS51168.2021.9636640
M3 - Conference contribution
T3 - IEEE International Conference on Intelligent Robots and Systems
SP - 9773
EP - 9779
BT - IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2021
ER -