LC-NAS: Latency Constrained Neural Architecture Search for Point Cloud Networks

Guohao Li*, Mengmeng Xu, Silvio Giancola, Ali Thabet, Bernard Ghanem

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations


Point cloud architecture design has become a crucial problem for deep learning in 3D. Several efforts have been made to manually design architectures targeting high accuracy in point cloud tasks such as classification, segmentation, and detection. Recent progress in automatic Neural Architecture Search (NAS) minimizes the human effort in network design and optimizes architectures for high performance. However, those efforts fail to consider crucial factors such as latency during inference, which is of high importance in time-critical and hardware-bounded applications like self-driving cars, robot navigation, and mobile applications. In this paper, we introduce a new NAS framework, dubbed LC-NAS, that searches for point cloud architectures constrained to a target latency. We implement a novel latency constraint formulation for the trade-off between accuracy and latency in our architecture search. Contrary to previous works, our latency loss enables us to find the best architecture with latency near a specific target value, which is crucial when the end task is to be deployed in a limited hardware setting. Extensive experiments show that LC-NAS is able to find state-of-the-art architectures for point cloud classification in ModelNet40 with a minimal computational cost. We also show how our searched architectures achieve any desired latency with a reasonably low drop in accuracy. Finally, we show how our searched architectures easily transfer to the part segmentation task on PartNet, where we achieve state-of-the-art results with significantly lower latency.

Original languageEnglish (US)
Title of host publicationProceedings - 2022 International Conference on 3D Vision, 3DV 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages11
ISBN (Electronic)9781665456708
StatePublished - 2022
Event10th International Conference on 3D Vision, 3DV 2022 - Prague, Czech Republic
Duration: Sep 12 2022Sep 15 2022

Publication series

NameProceedings - 2022 International Conference on 3D Vision, 3DV 2022


Conference10th International Conference on 3D Vision, 3DV 2022
Country/TerritoryCzech Republic

Bibliographical note

Funding Information:
We presented an automatic neural architecture search that considers the latency factor in the search. We designed a loss function that constrains the latency for a given hardware. We show with empirical results that our architectures LC-NAS reach the latency for which they have been designed on ModelNet10 and generalize on ModelNet40. Furthermore, we showed transfer capabilities of LC-NAS for part segmentation, showing state-of-the-art results on the PartNet benchmark. We envision LC-NAS to be used in time-constrained applications such as autonomous driv- ing, robotics, and embedded systems, where latency is of paramount importance for the fulfillment of the visual task. Acknowledgments. This work was supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research through the Visual Computing Center (VCC) funding.

Publisher Copyright:
© 2022 IEEE.


  • 3D Computer Vision
  • Graph Neural Networks
  • Neural Architecture Search

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing


Dive into the research topics of 'LC-NAS: Latency Constrained Neural Architecture Search for Point Cloud Networks'. Together they form a unique fingerprint.

Cite this