Accesso libero

Predicting Vehicle Pose in Six Degrees of Freedom from Single Image in Real-World Traffic Environments Using Deep Pretrained Convolutional Networks and Modified Centernet

Suresh Kolekar

Suresh Kolekar

  • Computer Engineering Department, Symbiosis Institute of Technology, Symbiosis International (Deemed) UniversityPune, India
  • Symbiosis Centre of Applied A.I. (SCAAI), Symbiosis International (Deemed) UniversityPune, India
Cerca questo autore su
Sciendo|Google Scholar
Kolekar, Suresh
, 
Shilpa Gite

Shilpa Gite

  • Symbiosis Centre of Applied A.I. (SCAAI), Symbiosis International (Deemed) UniversityPune, India
  • Artificial Intelligence and Machine Learning Department, Symbiosis Institute of Technology, Symbiosis International (Deemed) UniversityPune, India
Cerca questo autore su
Sciendo|Google Scholar
Gite, Shilpa
, 
Biswajeet Pradhan

Biswajeet Pradhan

Centre for Advanced Modelling and Geospatial Information Systems (CAMGIS), School of Civil and Environmental Engineering, Faculty of Engineering and I.T., University of Technology SydneyUltimo, Australia
Cerca questo autore su
Sciendo|Google Scholar
Pradhan, Biswajeet
 e 
Abdulla Alamri

Abdulla Alamri

Department of Geology and Geophysics, College of Science, King Saud UniversityRiyadh, Saudi Arabia
Cerca questo autore su
Sciendo|Google Scholar
Alamri, Abdulla
  
06 ago 2024
Predicting Vehicle Pose in Six Degrees of Freedom from Single Image in Real-World Traffic Environments Using Deep Pretrained Convolutional Networks and Modified Centernet's Cover Image
INFORMAZIONI SU QUESTO ARTICOLO
Articolo precedente
Articolo Successivo
Cita
Scarica la copertina
Categoria dell'articolo: Original Research Article
Pubblicato online: 06 ago 2024
Ricevuto: 18 apr 2024
DOI: https://doi.org/10.2478/ijssis-2024-0025
Parole chiave
© 2024 Suresh Kolekar et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1:

Schematic design of the proposed modified CenterNet model.
Schematic design of the proposed modified CenterNet model.

Figure 2:

ResNet50 architecture.
ResNet50 architecture.

Figure 3:

ResNext50 architecture.
ResNext50 architecture.

Figure 4:

Inception-ResNetV2 architecture.
Inception-ResNetV2 architecture.

Figure 5:

DenseNet201 architecture.
DenseNet201 architecture.

Figure 6:

Original images with ground truth labels.
Original images with ground truth labels.

Figure 7:

Comparative analysis of training loss.
Comparative analysis of training loss.

Figure 8:

Comparative analysis based on A3DP-Abs. A3DP-Abs, absolute translation thresholds.
Comparative analysis based on A3DP-Abs. A3DP-Abs, absolute translation thresholds.

Figure 9:

Comparative analysis based on A3DP-Rel. A3DP-Rel, relative translation thresholds.
Comparative analysis based on A3DP-Rel. A3DP-Rel, relative translation thresholds.

Performance evaluation of SOTA models based on A3DP-Rel

Model Mean C-l (loose criteria) C-S (strict criteria)
3D-RCNN (CVPR,18) 10.79 17.82 11.88
Direct-Based (CVPR,19) 11.49 17.82 11.88

Comparison of training loss

Model Regression loss Mask loss Total loss
Center-Inception-ResNetV2 0.85244 1.6311645 2.484085
Center-ResNet50 0.766475 0.666525 1.432925
Center-DenseNet201 0.450010 2.514673 2.964682
Center-ResNext50 0.466475 0.266525 0.733000

Performance evaluation of proposed models based on A3DP-Rel

Model Mean C-l (loose criteria) C-S (strict criteria)
Center-DenseNet201 11.82 23.89 10.85
Center-ResNet50 10.51 23.00 9.50
Center-Inception-ResNetV2 9.81 22.60 9.42
Center-ResNext50 9.61 22.17 9.04

Performance evaluation of SOTA models based on A3DP-Abs

Model Mean l (loose criteria) S (strict criteria)
3D-RCNN (CVPR,18) 16.44 29.70 19.80
Direct-Based (CVPR,19) 15.15 28.71 17.82

Performance evaluation of proposed models based on A3DP-Abs

Model Mean C-l (loose criteria) C-S (strict criteria)
Center-DenseNet201 39.92 52.284 44.68
Center-ResNet50 38.854 50.120 43.44
Center-Inception-ResNetV2 38.399 48.614 43.20
Center-ResNext50 37.06 47.027 41.52
Lingua:
Inglese
Frequenza di pubblicazione:
1 volte all'anno
Argomenti della rivista:
Ingegneria, Introduzioni e rassegna, Ingegneria, altro
Feed RSS della rivista