ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2021
DOI: 10.1109/icassp39728.2021.9414831
|View full text |Cite
|
Sign up to set email alerts
|

Branchy-GNN: A Device-Edge Co-Inference Framework for Efficient Point Cloud Processing

Abstract: The recent advancements of three-dimensional (3D) data acquisition devices have spurred a new breed of applications that rely on point cloud data processing. However, processing a large volume of point cloud data brings a significant workload on resource-constrained mobile devices, prohibiting from unleashing their full potentials. Built upon the emerging paradigm of device-edge co-inference, where an edge device extracts and transmits the intermediate feature to an edge server for further processing, we propo… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
17
0

Year Published

2021
2021
2025
2025

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 34 publications
(17 citation statements)
references
References 19 publications
0
17
0
Order By: Relevance
“…Well-being evaluates if GNNs are aligned to people's expectations regarding social good. 3 There is currently an urgent need to deploy GNNs on edge devices for real-time inference systems with limited processing power and memory resources (e.g., point cloud segmentation in autonomous vehicles [71]). As a representative sub-aspect of well-being, environmental well-being focuses on measuring the efficiency of GNNs.…”
Section: Trustworthy Gnnsmentioning
confidence: 99%
See 1 more Smart Citation
“…Well-being evaluates if GNNs are aligned to people's expectations regarding social good. 3 There is currently an urgent need to deploy GNNs on edge devices for real-time inference systems with limited processing power and memory resources (e.g., point cloud segmentation in autonomous vehicles [71]). As a representative sub-aspect of well-being, environmental well-being focuses on measuring the efficiency of GNNs.…”
Section: Trustworthy Gnnsmentioning
confidence: 99%
“…The explainability of GNNs refers to the ability to make the predictions of GNNs transparent and understandable. People will not fully trust GNNs if the predictions they make cannot be explained; this lack of trust will in turn limit their usage in crucial applications associated with fairness (e.g., credit risk prediction [64]), information security (e.g., chip design [124]) and life security (e.g., autonomous vehicles [71], protein structure prediction [125]). Consequently, building trustworthy GNNs requires insights into why GNNs make particular predictions, which has driven an increase in research into the interpretability and explainability of GNNs.…”
Section: Explainability Of Gnnsmentioning
confidence: 99%
“…• Split inference: Feature extraction [79,80,[80][81][82][83], importance-aware RRM [80,[84][85][86][87][88], Split-Net approach [89][90][91][92].…”
Section: M2m Semcommentioning
confidence: 99%
“…Modern feature extraction exploits the powerful representation capability of neural networks and rich training data. Such a feature-extraction model can be implemented multi-layer perceptrons (MLPs) for a general purpose, CNNs for visual data [80,81], and RNNs for time-series data [82] or leverage the emerging graph neural networks to improve inference performance with point cloud and non-Euclidean data [83].…”
Section: Effectiveness Encoding and Transmission Formentioning
confidence: 99%
“…There are several works solving some tasks in point cloud using GNNs. J. Shao et al proposed Branchy-GNN for efficient point cloud processing [74]. Branchy-GNN uses branch network and source channel coding to reduce the computational cost and intermediate feature transmission overhead on the device, respectively.…”
Section: F Point Cloudmentioning
confidence: 99%