Bottleneck Identification to Semantic Segmentation of Industrial 3D Point Cloud Scene via Deep Learning

Romain Cazorla; Line Poinel; Panagiotis Papadakis; Cédric Buche

IJCAI 2021poster6 citations

Bottleneck Identification to Semantic Segmentation of Industrial 3D Point Cloud Scene via Deep Learning

Romain Cazorla, Line Poinel, Panagiotis Papadakis, Cédric Buche

Abstract

Point cloud acquisition techniques are an essential tool for the digitization of industrial plants, yet the bulk of a designer's work remains manual. A first step to automatize drawing generation is to extract the semantics of the point cloud. Towards this goal, we investigate the use of deep learning to semantically segment oil and gas industrial scenes. We focus on domain characteristics such as high variation of object size, increased concavity and lack of annotated data, which hampers the use of conventional approaches. To address these issues, we advocate the use of synthetic data, adaptive downsampling and context sharing.

Computer Vision: 2D and 3D Computer VisionMachine Learning: Deep LearningMachine Learning Applications: Applications of Supervised LearningComputer Vision: Recognition: Detection, Categorization, Indexing, Matching, Retrieval, Semantic Interpretation

BibTeX

@inproceedings{ijcai2021p670,
  title     = {Bottleneck Identification to Semantic Segmentation of Industrial 3D Point Cloud Scene via Deep Learning},
  author    = {Cazorla, Romain and Poinel, Line and Papadakis, Panagiotis and Buche, Cédric},
  booktitle = {Proceedings of the Thirtieth International Joint Conference on
               Artificial Intelligence, {IJCAI-21}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Zhi-Hua Zhou},
  pages     = {4877--4878},
  year      = {2021},
  month     = {8},
  note      = {Doctoral Consortium},
  doi       = {10.24963/ijcai.2021/670},
  url       = {https://doi.org/10.24963/ijcai.2021/670},
}