Conference paper · 2021 · Open access
RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices
Proceedings of the AAAI Conference on Artificial Intelligence, 35(10), 9179–9187 · Published 18 May 2021 · DOI 10.1609/aaai.v35i10.17108
In brief
“RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices” is a 2021 conference paper by Wei Niu and 9 co-authors, published in Proceedings of the AAAI Conference on Artificial Intelligence. OpenAlex counts 15 citations as of 19 November 2026. It is open access: the full text is free to read at Proceedings of the AAAI Conference on Artificial Intelligence.
- Citations
- 15
- Most cited in
- 2022 (5)
- Field-weighted impact
- 0.62
- References
- 50
- Authors
- 10
Abstract
Mobile devices are becoming an important carrier for deep learning tasks, as they are being equipped with powerful, high-end mobile CPUs and GPUs. However, it is still a challenging task to execute 3D Convolutional Neural Networks (CNNs) targeting for real-time performance, besides high inference accuracy. The reason is more complex model structure and higher model dimensionality overwhelm the available computation/storage resources on mobile devices. A natural way may be turning to deep learning weight pruning techniques. However, the direct generalization of existing 2D CNN weight pruning methods to 3D CNNs is not ideal for fully exploiting mobile parallelism while achieving high inference accuracy. This paper proposes RT3D, a model compression and mobile acceleration framework for 3D CNNs, seamlessly integrating neural network weight pruning and compiler code generation techniques. We propose and investigate two structured sparsity schemes i.e., the vanilla structured sparsity and kernel group structured (KGS) sparsity that are mobile acceleration friendly. The vanilla sparsity removes whole kernel groups, while KGS sparsity is a more fine-grained structured sparsity that enjoys higher flexibility while exploiting full on-device parallelism. We propose a reweighted regularization pruning algorithm to achieve the proposed sparsity schemes. The inference time speedup due to sparsity is approaching the pruning rate of the whole model FLOPs (floating point operations). RT3D demonstrates up to 29.1x speedup in end-to-end inference time comparing with current mobile frameworks supporting 3D CNNs, with moderate 1%~1.5% accuracy loss. The end-to-end inference time for 16 video frames could be within 150 ms, when executing representative C3D and R(2+1)D models on a cellphone. For the first time, real-time execution of 3D CNNs is achieved on off-the-shelf mobiles.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Niu, W., Sun, M., Li, Z., Chen, J.-A., Guan, J., Shen, X., Wang, Y., Liu, S., Lin, X., & Ren, B. (2021). RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices. Proceedings of the AAAI Conference on Artificial Intelligence, 35(10), 9179–9187. https://doi.org/10.1609/aaai.v35i10.17108
MLA 9
Niu, Wei, et al. “RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices.” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 10, 2021, pp. 9179–9187. https://doi.org/10.1609/aaai.v35i10.17108.
Chicago
Niu, Wei, Mengshu Sun, Zhengang Li, Jou-An Chen, Jiexiong Guan, Xipeng Shen, Yanzhi Wang, Sijia Liu, Xue Lin, and Bin Ren. 2021. “RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices.” Proceedings of the AAAI Conference on Artificial Intelligence 35 (10): 9179–9187. https://doi.org/10.1609/aaai.v35i10.17108.
Harvard
Niu, W. et al. (2021) ‘RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices’, Proceedings of the AAAI Conference on Artificial Intelligence, 35(10), pp. 9179–9187. doi:10.1609/aaai.v35i10.17108.
IEEE
W. Niu et al., “RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 10, pp. 9179–9187, 2021, doi: 10.1609/aaai.v35i10.17108.
BibTeX
@inproceedings{niu2021achieving,
title = {{RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices}},
author = {Niu, Wei and Sun, Mengshu and Li, Zhengang and Chen, Jou-An and Guan, Jiexiong and Shen, Xipeng and Wang, Yanzhi and Liu, Sijia and Lin, Xue and Ren, Bin},
booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2021},
volume = {35},
number = {10},
pages = {9179--9187},
doi = {10.1609/aaai.v35i10.17108},
url = {https://doi.org/10.1609/aaai.v35i10.17108}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Citations by year
| Year | Citations | Share of the best year |
|---|---|---|
| 2021 | 1 | |
| 2022 | 5 | |
| 2023 | 2 | |
| 2024 | 2 | |
| 2025 | 3 | |
| 2026 | 2 |
Its field-weighted citation impact is 0.62: it has been cited 0.62 times as often as the average work of the same field, type and year (1.00 is the world average).
Topic and keywords
Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- 3D convolutional neural networks
- real-time inference
- mobile devices
- weight pruning
- structured sparsity
- model compression
- compiler code generation
- C3d
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Niu, W., Sun, M., Li, Z., Chen, J.-A., Guan, J., Shen, X., Wang, Y., Liu, S., Lin, X., & Ren, B. (2021). RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices. Proceedings of the AAAI Conference on Artificial Intelligence, 35(10), 9179–9187. https://doi.org/10.1609/aaai.v35i10.17108
How many times has this paper been cited?
15 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at Proceedings of the AAAI Conference on Artificial Intelligence: sign in and open the full text from this page.
Who wrote this paper?
Wei Niu, Mengshu Sun, Zhengang Li, Jou-An Chen, Jiexiong Guan, Xipeng Shen, Yanzhi Wang, Sijia Liu, Xue Lin, Bin Ren.
What is the DOI of this paper?
10.1609/aaai.v35i10.17108, permanently at https://doi.org/10.1609/aaai.v35i10.17108. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.