This paper addresses the challenge of azimuth estimation in the context of car pose estimation. Our research utilizes the PASCAL3D+ dataset, which offers a diverse range of object categories, including cars, with annotated azimuth estimations for each photograph. We introduce two architectures that approach azimuth estimation as a regression problem, each employing a deep convolutional neural network (DCNN) backbone but diverging in their output definition strategies. The first architecture employs a sin-cos representation of the car’s azimuth, while the second utilizes two directional discriminators, distinguishing between front/rear and left/right views of the vehicle. Our comparative analysis reveals that both architectures demonstrate near-identical performance levels on the PASCAL3D+ validation set, achieving a median error of 3.5◦, which is a significant advancement in the state of the art. The minimal performance disparity between the two methods highlights their individual strengths while also underscoring the similarity in their practical efficacy. This study not only proposes effective solutions for accurate azimuth estimation but also contributes to the broader understanding of pose estimation challenges in automotive contexts. The code is available at https://github.com/vani-or/car_pose_estimation.

Orlov, I., Buzzelli, M., Schettini, R. (2024). Vehicle Pose Estimation: Exploring Angular Representations. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - (Volume 2) (pp.853-860). Science and Technology Publications, Lda [10.5220/0012574300003660].

Vehicle Pose Estimation: Exploring Angular Representations

Orlov, I;Buzzelli, M
;
Schettini, R
2024

Abstract

This paper addresses the challenge of azimuth estimation in the context of car pose estimation. Our research utilizes the PASCAL3D+ dataset, which offers a diverse range of object categories, including cars, with annotated azimuth estimations for each photograph. We introduce two architectures that approach azimuth estimation as a regression problem, each employing a deep convolutional neural network (DCNN) backbone but diverging in their output definition strategies. The first architecture employs a sin-cos representation of the car’s azimuth, while the second utilizes two directional discriminators, distinguishing between front/rear and left/right views of the vehicle. Our comparative analysis reveals that both architectures demonstrate near-identical performance levels on the PASCAL3D+ validation set, achieving a median error of 3.5◦, which is a significant advancement in the state of the art. The minimal performance disparity between the two methods highlights their individual strengths while also underscoring the similarity in their practical efficacy. This study not only proposes effective solutions for accurate azimuth estimation but also contributes to the broader understanding of pose estimation challenges in automotive contexts. The code is available at https://github.com/vani-or/car_pose_estimation.
slide + paper
Angular Regression; Car Azimuth Estimation; PASCAL3D+; Vehicle Pose Recognition; Viewpoint Estimation;
English
19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISAPP 2024) - February 27-29, 2024
2024
Radeva, P; Furnari, A; Bouatouch, K; Sousa, A
Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - (Volume 2)
9789897586798
2024
2
853
860
https://www.scitepress.org/ProceedingsDetails.aspx?ID=z7/YTqfSA1o=&t=1
open
Orlov, I., Buzzelli, M., Schettini, R. (2024). Vehicle Pose Estimation: Exploring Angular Representations. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - (Volume 2) (pp.853-860). Science and Technology Publications, Lda [10.5220/0012574300003660].
File in questo prodotto:
File Dimensione Formato  
Orlov-2024-VISAPP-VoR.pdf

accesso aperto

Tipologia di allegato: Publisher’s Version (Version of Record, VoR)
Licenza: Creative Commons
Dimensione 2.14 MB
Formato Adobe PDF
2.14 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/463619
Citazioni
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
Social impact