TY - GEN
T1 - Out-of-Distribution Detection for Adaptive Computer Vision
AU - Kristoffersson Lind, Simon
AU - Triebel, Rudolph
AU - Nardi, Luigi
AU - Krueger, Volker
N1 - Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2023
Y1 - 2023
N2 - It is well known that computer vision can be unreliable when faced with previously unseen imaging conditions. This paper proposes a method to adapt camera parameters according to a normalizing flow-based out-of-distibution detector. A small-scale study is conducted which shows that adapting camera parameters according to this out-of-distibution detector leads to an average increase of 3 to 4% points in mAP, mAR and F1 performance metrics of a YOLOv4 object detector. As a secondary result, this paper also shows that it is possible to train a normalizing flow model for out-of-distribution detection on the COCO dataset, which is larger and more diverse than most benchmarks for out-of-distibution detectors.
AB - It is well known that computer vision can be unreliable when faced with previously unseen imaging conditions. This paper proposes a method to adapt camera parameters according to a normalizing flow-based out-of-distibution detector. A small-scale study is conducted which shows that adapting camera parameters according to this out-of-distibution detector leads to an average increase of 3 to 4% points in mAP, mAR and F1 performance metrics of a YOLOv4 object detector. As a secondary result, this paper also shows that it is possible to train a normalizing flow model for out-of-distribution detection on the COCO dataset, which is larger and more diverse than most benchmarks for out-of-distibution detectors.
KW - Autonomous Systems
KW - Normalizing Flows
KW - Object Detection
KW - Out-of-Distribution Detection
UR - http://www.scopus.com/inward/record.url?scp=85161436449&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-31438-4_21
DO - 10.1007/978-3-031-31438-4_21
M3 - Conference contribution
AN - SCOPUS:85161436449
SN - 9783031314377
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 311
EP - 325
BT - Image Analysis - 23rd Scandinavian Conference, SCIA 2023, Proceedings
A2 - Gade, Rikke
A2 - Felsberg, Michael
A2 - Kämäräinen, Joni-Kristian
PB - Springer Science and Business Media Deutschland GmbH
T2 - 23nd Scandinavian Conference on Image Analysis, SCIA 2023
Y2 - 18 April 2023 through 21 April 2023
ER -