A carregar...
Publicação
Assessing the 3D Position of a Car with a Single 2D Camera Using Siamese Networks
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 3.11 MB | Adobe PDF |
Orientador(es)
Resumo(s)
Using computer vision for the classification of an object’s 3D
position using a 2D camera is a topic that has received some attention
from researchers over the years. Visual data is interpreted by the computer
to recognize the objects found. In addition, it is possible to infer
their orientation, evaluating their spatial arrangement, rotation, or alignment
in the scene. The work presented in this paper describes the training
and selection of a siamese neural network for classifying the 3D orientation
of cars using 2D images. The neural network is composed of an
initial phase for feature selection through convolutional neural networks
followed by a dense layer for embedding generation. For feature selection,
four architectures were tested: VGG16, VGG19, ResNet18 and ResNet50.
The best result of 95.8% accuracy was obtained with the VGG16 and
input images preprocessed for background removal.
Descrição
Palavras-chave
Computer Vision Maintenance support Siamese networks Object Orientation
Contexto Educativo
Citação
Yahia, Youssef Bel Haj; Lopes, Júlio Castro; Bezerra, Eduardo; Rodrigues, Pedro João; Lopes, Rui Pedro (2024). Assessing the 3D Position of a Car with a Single 2D Camera Using Siamese Networks. In 3rd International Conference on Optimization, Learning Algorithms and Applications (OL2A 2023). Cham: Springer Nature, Vol. 2, p. 93–107. ISBN 978-3-031-53035-7
Editora
Springer Nature