Type of Publication

Journal Articles

Date:

5 /

2023

Status

Published

DOI:

10.1109/TITS.2023.3268578

Probabilistic Approach for Road-Users Detection

Featured in:

IEEE Transactions on Intelligent Transportation Systems

Authors:

Gledson Melotti, Weihao Lu, Pedro Conde, Dezong Zhao, Alireza Asvadi, Nuno Gonçalves and Cristiano Premebida

Abstract

Object detection in autonomous driving applications implies that the detection and tracking of semantic objects are commonly native to urban driving environments, as pedestrians and vehicles. One of the major challenges in state-of-the-art deep-learning based object detection are false positives which oc- cur with overconfident scores. This is highly undesir- able in autonomous driving and other critical robotic- perception domains because of safety concerns. This paper proposes an approach to alleviate the problem of overconfident predictions by introducing a novel probabilistic layer to deep object detection networks in testing. The suggested approach avoids the tradi- tional Sigmoid or Softmax prediction layer which often produces overconfident predictions. It is demonstrated that the proposed technique reduces overconfidence in the false positives without degrading the perfor- mance on the true positives. The approach is validated on the 2D-KITTI objection detection through the YOLOV4 and SECOND (Lidar-based detector). The proposed approach enables interpretable probabilistic predictions without the requirement of re-training the network and therefore is very practical.

Citation
Gledson Melotti, Weihao Lu, Pedro Conde, Dezong Zhao, Alireza Asvadi, Nuno Gonçalves and Cristiano Premebida (2023). Probabilistic approach for road-users detection. IEEE Transactions on Intelligent Transportation Systems. DOI: 10.1109/TITS.2023.3268578.

Related Content

Researcher Coordinator, VIS TEAM Leader
PhD Student
No tagged content to show
No tagged content to show
No tagged content to show

RECENT PUBLICATIONS

VOIDFace: A Privacy-Preserving Multi-Network Face Recognition With Enhanced Security

Authors: Ajnas Muhammed; Iurii Medvedev; Nuno Gonçalves
Featured in: IEEE International Joint Conference on Biometrics (IJCB 2025)

Using Benford’s Law for Deepfake Detection

Authors: Miguel Leão; Nuno Gonçalves
Featured in: RECPAD - 30th Portuguese Conference on Pattern Recognition. 2024, Covilhã, Portugal

Part I – Proceedings of the 12th Iberian Conference on Pattern Recognition and Image Analysis

Authors: Nuno Gonçalves; Hélder P. Oliveira; Joan Andreu Sánchez
Featured in: 12th Iberian Conference on Pattern Recognition and Image Analysis (IbPRIA 2025)

suggested news

Paper accepted to IJCB 2025
Prof. Nuno and VIS Team successfully organizes IbPRIA...
Four papers presented @ IbPRIA 2025

RECENT PROJECTS

FACING2 – Face Image Understanding
VISUAL-ID – Unique Visual Identities in Graphics, Images and Faces
UniqueMark

Institute of Systems and Robotics Department of Electrical and Computers Engineering University of Coimbra