TY - GEN
T1 - A Flow-Based Credibility Metric for Safety-Critical Pedestrian Detection
AU - Lyssenko, Maria
AU - Gladisch, Christoph
AU - Heinzemann, Christian
AU - Woehrle, Matthias
AU - Triebel, Rudolph
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
PY - 2024
Y1 - 2024
N2 - Safety is of utmost importance for perception in automated driving (AD). However, a prime safety concern in state-of-the-art object detection is that standard evaluation schemes utilize safety-agnostic metrics to argue for sufficient detection performance. Hence, it is imperative to leverage supplementary domain knowledge to accentuate safety-critical misdetections during evaluation tasks. To tackle the underspecification, this paper introduces a novel credibility metric, called c-flow, for pedestrian bounding boxes. To this end, c-flow relies on a complementary optical flow signal from image sequences and enhances the analyses of safety-critical misdetections without requiring additional labels. We implement and evaluate c-flow with a state-of-the-art pedestrian detector on a large AD dataset. Our analysis demonstrates that c-flow allows developers to identify safety-critical misdetections.
AB - Safety is of utmost importance for perception in automated driving (AD). However, a prime safety concern in state-of-the-art object detection is that standard evaluation schemes utilize safety-agnostic metrics to argue for sufficient detection performance. Hence, it is imperative to leverage supplementary domain knowledge to accentuate safety-critical misdetections during evaluation tasks. To tackle the underspecification, this paper introduces a novel credibility metric, called c-flow, for pedestrian bounding boxes. To this end, c-flow relies on a complementary optical flow signal from image sequences and enhances the analyses of safety-critical misdetections without requiring additional labels. We implement and evaluate c-flow with a state-of-the-art pedestrian detector on a large AD dataset. Our analysis demonstrates that c-flow allows developers to identify safety-critical misdetections.
KW - Optical Flow
KW - Safe Perception in AD
KW - Verification & Validation (V & V)
UR - http://www.scopus.com/inward/record.url?scp=85204622137&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-68738-9_26
DO - 10.1007/978-3-031-68738-9_26
M3 - Conference contribution
AN - SCOPUS:85204622137
SN - 9783031687372
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 335
EP - 350
BT - Computer Safety, Reliability, and Security. SAFECOMP 2024 Workshops - DECSoS, SASSUR, TOASTS, and WAISE, Proceedings
A2 - Ceccarelli, Andrea
A2 - Bondavalli, Andrea
A2 - Trapp, Mario
A2 - Schoitsch, Erwin
A2 - Gallina, Barbara
A2 - Bitsch, Friedemann
PB - Springer Science and Business Media Deutschland GmbH
T2 - 19th Workshop on Dependable Smart Embedded and Cyber-Physical Systems and Systems-of-Systems, DECSoS 2024, 11th International Workshop on Next Generation of System Assurance Approaches for Critical Systems, SASSUR 2024, Towards A Safer Systems architecture Through Security, TOASTS 2024 and 7th International Workshop on Artificial Intelligence Safety Engineering, WAISE 2024 held in conjunction with the 43rd International Conference on Computer Safety, Reliability, and Security, SAFECOMP 2024
Y2 - 17 September 2024 through 17 September 2024
ER -