VoxRec: Hybrid convolutional neural network for active 3D object recognition

Ahmad Karambakhsh, Bin Sheng, Ping Li, Po Yang, Younhyun Jung, David Dagan Feng

Research output: Journal article publicationJournal articleAcademic researchpeer-review

4 Citations (Scopus)


Deep Neural Network methods have been used to a variety of challenges in automatic 3D recognition. Although discovered techniques provide many advantages in comparison with conventional methods, they still suffer from different drawbacks, e.g., a large number of pre-processing stages and time-consuming training. In this paper, an innovative approach has been suggested for recognizing 3D models. It contains encoding 3D point clouds, surface normal, and surface curvature, merge them to provide more effective input data, and train it via a deep convolutional neural network on Shapenetcore dataset. We also proposed a similar method for 3D segmentation using Octree coding method. Finally, comparing the accuracy with some of the state-of-the-art demonstrates the effectiveness of our proposed method.

Original languageEnglish
Pages (from-to)70969-70980
Number of pages12
JournalIEEE Access
Publication statusPublished - Apr 2020


  • multi-layer neural network
  • Object recognition
  • octrees
  • recurrent neural networks

ASJC Scopus subject areas

  • General Computer Science
  • General Materials Science
  • General Engineering


Dive into the research topics of 'VoxRec: Hybrid convolutional neural network for active 3D object recognition'. Together they form a unique fingerprint.

Cite this