CAPS-BQ: Uma Abordagem Compacta Baseada em Atenção para Segmentação de Partes em Objetos 3D
Resumo
O progresso tecnológico impulsiona o surgimento de sensores de alta precisão para representação de escaneamento tridimensional, como o LiDAR. A Segmentação de instâncias de nuvens de pontos, originárias de sensores desta natureza, são tarefas fundamentais no processamento e análise desses dados, trazendo informações valiosas e úteis para tomadas de decisão. Esse estudo propõe uma abordagem de rede neural leve, baseada em atenção, para segmentação de partes de nuvens de pontos, usando como critério seletor de pontos um algoritmo de busca baseada em raio. Utilizando um conjunto de dados amplamente difundido na literatura, essa pesquisa propõe o desenvolvimento de um modelo leve, robusto e computacionalmente eficiente. O modelo proposto alcançou um resultado competitivo de 84,26% de mIoU, inferindo aproximadamente 45,56% mais rápido em comparação com o seu melhor competidor sobre 2874 nuvens de pontos, demonstrando eficiência computacional. A análise comparativa foi realizada considerando os dois modelos da APES, Local e Global, a rede PartField e a consolidada PointNet, demonstrando a eficácia da arquitetura proposta em manter alto desempenho com um tempo de inferência otimizado.Referências
Y. Guo, H. Wang, Q. Hu, H. Liu, L. Liu, and M. Bennamoun, “Deep learning for 3d point clouds: A survey,” IEEE transactions on pattern analysis and machine intelligence, vol. 43, no. 12, pp. 4338–4364, 2020.
H. Ran, J. Liu, and C. Wang, “Surface representation for point clouds,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 18 942–18 952.
R. Richter and J. D¨ollner, “Concepts and techniques for integration, analysis and visualization of massive 3d point clouds,” Computers, Environment and Urban Systems, vol. 45, pp. 114–124, 2014.
C. R. Qi, H. Su, K. Mo, and L. J. Guibas, “Pointnet: Deep learning on point sets for 3d classification and segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 652–660.
M. Liu, M. A. Uy, D. Xiang, H. Su, S. Fidler, N. Sharp, and J. Gao, “Partfield: Learning 3d feature fields for part segmentation and beyond,” arXiv preprint arXiv:2504.11451, 2025.
Y. Yang, Y. Huang, Y.-C. Guo, L. Lu, X. Wu, E. Y. Lam, Y.-P. Cao, and X. Liu, “Sampart3d: Segment any part in 3d objects,” arXiv preprint arXiv:2411.07184, 2024.
C. Wu, J. Zheng, J. Pfrommer, and J. Beyerer, “Attention-based point cloud edge sampling,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5333–5343.
L. Yi, V. G. Kim, D. Ceylan, I.-C. Shen, M. Yan, H. Su, C. Lu, Q. Huang, A. Sheffer, and L. Guibas, “A scalable active framework for region annotation in 3d shape collections,” ACM Transactions on Graphics (ToG), vol. 35, no. 6, pp. 1–12, 2016.
L. E. Peterson, “K-nearest neighbor,” Scholarpedia, vol. 4, no. 2, p. 1883, 2009.
R. Meyes, M. Lu, C. W. De Puiseau, and T. Meisen, “Ablation studies in artificial neural networks,” arXiv preprint arXiv:1901.08644, 2019.
C. R. Qi, L. Yi, H. Su, and L. J. Guibas, “Pointnet++: Deep hierarchical feature learning on point sets in a metric space,” Advances in neural information processing systems, vol. 30, 2017.
J. Cen, Z. Zhou, J. Fang, W. Shen, L. Xie, D. Jiang, X. Zhang, Q. Tian et al., “Segment anything in 3d with nerfs,” Advances in Neural Information Processing Systems, vol. 36, pp. 25 971–25 990, 2023.
Y. He, H. Yu, X. Liu, Z. Yang, W. Sun, S. Anwar, and A. Mian, “Deep learning based 3d segmentation in computer vision: A survey,” Information Fusion, vol. 115, p. 102722, 2025.
D. Parekh, N. Poddar, A. Rajpurkar, M. Chahal, N. Kumar, G. P. Joshi, and W. Cho, “A review on autonomous vehicles: Progress, methods and challenges,” Electronics, vol. 11, no. 14, p. 2162, 2022.
H. Ran, J. Liu, and C. Wang, “Surface representation for point clouds,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 18 942–18 952.
R. Richter and J. D¨ollner, “Concepts and techniques for integration, analysis and visualization of massive 3d point clouds,” Computers, Environment and Urban Systems, vol. 45, pp. 114–124, 2014.
C. R. Qi, H. Su, K. Mo, and L. J. Guibas, “Pointnet: Deep learning on point sets for 3d classification and segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 652–660.
M. Liu, M. A. Uy, D. Xiang, H. Su, S. Fidler, N. Sharp, and J. Gao, “Partfield: Learning 3d feature fields for part segmentation and beyond,” arXiv preprint arXiv:2504.11451, 2025.
Y. Yang, Y. Huang, Y.-C. Guo, L. Lu, X. Wu, E. Y. Lam, Y.-P. Cao, and X. Liu, “Sampart3d: Segment any part in 3d objects,” arXiv preprint arXiv:2411.07184, 2024.
C. Wu, J. Zheng, J. Pfrommer, and J. Beyerer, “Attention-based point cloud edge sampling,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5333–5343.
L. Yi, V. G. Kim, D. Ceylan, I.-C. Shen, M. Yan, H. Su, C. Lu, Q. Huang, A. Sheffer, and L. Guibas, “A scalable active framework for region annotation in 3d shape collections,” ACM Transactions on Graphics (ToG), vol. 35, no. 6, pp. 1–12, 2016.
L. E. Peterson, “K-nearest neighbor,” Scholarpedia, vol. 4, no. 2, p. 1883, 2009.
R. Meyes, M. Lu, C. W. De Puiseau, and T. Meisen, “Ablation studies in artificial neural networks,” arXiv preprint arXiv:1901.08644, 2019.
C. R. Qi, L. Yi, H. Su, and L. J. Guibas, “Pointnet++: Deep hierarchical feature learning on point sets in a metric space,” Advances in neural information processing systems, vol. 30, 2017.
J. Cen, Z. Zhou, J. Fang, W. Shen, L. Xie, D. Jiang, X. Zhang, Q. Tian et al., “Segment anything in 3d with nerfs,” Advances in Neural Information Processing Systems, vol. 36, pp. 25 971–25 990, 2023.
Y. He, H. Yu, X. Liu, Z. Yang, W. Sun, S. Anwar, and A. Mian, “Deep learning based 3d segmentation in computer vision: A survey,” Information Fusion, vol. 115, p. 102722, 2025.
D. Parekh, N. Poddar, A. Rajpurkar, M. Chahal, N. Kumar, G. P. Joshi, and W. Cho, “A review on autonomous vehicles: Progress, methods and challenges,” Electronics, vol. 11, no. 14, p. 2162, 2022.
Publicado
30/09/2025
Como Citar
BARROSO, Calleo B.; MOREIRA, Hector L. M.; SILVA, Francisco Hércules dos S.; SANTOS, José Daniel de A.; REBOUÇAS FILHO, Pedro P..
CAPS-BQ: Uma Abordagem Compacta Baseada em Atenção para Segmentação de Partes em Objetos 3D. In: WORKSHOP DE TRABALHOS EM ANDAMENTO - CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI), 38. , 2025, Salvador/BA.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 120-125.
