A Scalable Real-time Semantic Segmentation Network for Autonomous Driving

dc.contributor.authorLiu, Jun
dc.contributor.authorWu, Chao
dc.contributor.authorYuan, Geng
dc.contributor.authorNiu, Wei
dc.contributor.authorZhang, Wenbin
dc.contributor.authorSong, Houbing
dc.date.accessioned2023-11-07T21:01:01Z
dc.date.available2023-11-07T21:01:01Z
dc.date.issued2023-10-29
dc.descriptionMM '23: The 31st ACM International Conference on Multimedia; Ottawa, ON, Canada; 29 October 2023en_US
dc.description.abstractComputational resources are limited on real-time embedded devices, so the available computing cost of deployment on the target platform must be considered. We develop a feature extraction module based on the MobileNet backbone that can be adjusted in terms of computational complexity and capacity using the depth multiplier parameter, classifier depth, and kernel depth. These three parameters allow us to control the count of channels within the network, effectively managing the model's capacity and computational requirements. To achieve semantic segmentation, we incorporate additional components, such as an extension module. This extension module typically includes 1x1 pointwise convolutional layers for pixel-level classification and a transposed convolutional layer for upsampling the output to the original input image size. By combining the feature extraction module with this extension module, we can create a complete architecture capable of performing semantic segmentation tasks. The feature extraction module provides the initial feature extraction and the extension module adds the necessary components for accurate pixel-wise classification and upsampling. Compared to Hardware-aware Neural Architecture Search (NAS), pruning, runtime pruning, and knowledge distillation methods, our model has several advantages in terms of modular design, structural controllability, ease of implementation, and cost-effectiveness. Our computational efficiency, as measured by FLOPS, is highly competitive. Our method is distinguished by solving the problem of MobileNet's inability to adjust the size and number of convolution kernels. It achieves this through adaptable parameter tuning, including MobileNet's depth multiplier, the kernel size in the FCN head's Separable Convolution layer, and the depth of the first Point-wise Convolution layer. These adjustments are customized to match hardware's max multiply-accumulates (MACs), optimizing network capacity and maximizing resource utilization.en_US
dc.description.sponsorshipWe sincerely thank the NVIDIA Corporation for their technical support throughout the research process.en_US
dc.description.urihttps://dl.acm.org/doi/abs/10.1145/3606042.3616451en_US
dc.format.extent10 pagesen_US
dc.genreconference papers and proceedingsen_US
dc.identifierdoi:10.13016/m2qm8z-qfsa
dc.identifier.citationLiu, Jun, Chao Wu, Geng Yuan, Wei Niu, Wenbin Zhang, and Houbing Herbert Song. “A Scalable Real-Time Semantic Segmentation Network for Autonomous Driving.” In Proceedings of the 2023 Workshop on Advanced Multimedia Computing for Smart Manufacturing and Engineering, 3–12. AMC-SME ’23. New York, NY, USA: Association for Computing Machinery, 2023. https://doi.org/10.1145/3606042.3616451.en_US
dc.identifier.urihttp://hdl.handle.net/11603/30583
dc.language.isoen_USen_US
dc.publisherACMen_US
dc.relation.isAvailableAtThe University of Maryland, Baltimore County (UMBC)
dc.relation.ispartofUMBC Information Systems Department Collection
dc.relation.ispartofUMBC Faculty Collection
dc.rightsThis item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.en_US
dc.rightsAttribution 4.0 International (CC BY 4.0 DEED)*
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/*
dc.titleA Scalable Real-time Semantic Segmentation Network for Autonomous Drivingen_US
dc.typeTexten_US
dcterms.creatorhttps://orcid.org/0000-0003-2631-9223en_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
3606042.3616451.pdf
Size:
1.09 MB
Format:
Adobe Portable Document Format
Description:

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
2.56 KB
Format:
Item-specific license agreed upon to submission
Description: