publications

2024

1. arXiv

   

   Real-Time Human Action Recognition on Embedded Platforms

   Ruiqi Wang, Zichen Wang, Peiqi Gao, Mingzhen Li, Jaehwan Jeong, Yihang Xu, Yejin Lee, Carolyn M. Baum, Lisa Tabor Connor, and Chenyang Lu

   arXiv preprint arXiv:2409.05662, 2024

   Abs arXiv

   With advancements in computer vision and deep learning, video-based human action recognition (HAR) has become practical. However, due to the complexity of the computation pipeline, running HAR on live video streams incurs excessive delays on embedded platforms. This work tackles the real-time performance challenges of HAR with four contributions: 1) an experimental study identifying a standard Optical Flow (OF) extraction technique as the latency bottleneck in a state-of-the-art HAR pipeline, 2) an exploration of the latency-accuracy tradeoff between the standard and deep learning approaches to OF extraction, which highlights the need for a novel, efficient motion feature extractor, 3) the design of Integrated Motion Feature Extractor (IMFE), a novel single-shot neural network architecture for motion feature extraction with drastic improvement in latency, 4) the development of RT-HARE, a real-time HAR system tailored for embedded platforms. Experimental results on an Nvidia Jetson Xavier NX platform demonstrated that RT-HARE realizes real-time HAR at a video frame rate of 30 frames per second while delivering high levels of recognition accuracy.

2. SEC 2024

   

   Optimizing Edge Offloading Decisions for Object Detection

   Jiaming Qiu, Ruiqi Wang, Brooks Hu, Roch Guérin, and Chenyang Lu

   In 2024 IEEE/ACM Symposium on Edge Computing (SEC) , 2024

   Abs arXiv Bib HTML Code

   Recent advances in machine learning and hardware have produced embedded devices capable of performing real-time object detection with commendable accuracy. We consider a scenario in which embedded devices rely on an onboard object detector, but have the option to offload detection to a more powerful edge server when local accuracy is deemed too low. Resource constraints, however, limit the number of images that can be offloaded to the edge. Our goal is to identify which images to offload to maximize overall detection accuracy under those constraints. To that end, the paper introduces a reward metric designed to quantify potential accuracy improvements from offloading individual images, and proposes an efficient approach to make offloading decisions by estimating this reward based only on local detection results. The approach is computationally frugal enough to run on embedded devices, and empirical findings indicate that it outperforms existing alternatives in improving detection accuracy even when the fraction of offloaded images is small. Code for the paper’s solution is available at https://github.com/qiujiaming315/edgeml-object-detection.

   @inproceedings{qiu2024optimizing,
     author = {Qiu, Jiaming and Wang, Ruiqi and Hu, Brooks and Guérin, Roch and Lu, Chenyang},
     booktitle = {2024 IEEE/ACM Symposium on Edge Computing (SEC)},
     title = {Optimizing Edge Offloading Decisions for Object Detection},
     year = {2024},
     volume = {},
     number = {},
     pages = {164-177},
     organization = {IEEE},
     keywords = {Accuracy;Correlation;Image edge detection;System performance;Estimation;Detectors;Object detection;Real-time systems;Servers;Edge computing;edge AI;object detection;embedded machine learning;distributed computing},
     doi = {10.1109/SEC62691.2024.00021},
   }

3. Infection Control
   & Hospital
   Epidemiology

   

   Leveraging Bluetooth low-energy technology to improve contact tracing among healthcare personnel in hospital setting during the coronavirus disease 2019 (COVID-19) pandemic

   M. Cristina Vazquez Guillamet, Ashraf Rjob, Jingwen Zhang, Ruixuan Dai, Ruiqi Wang, Christopher Damulira, Reshad Hamauon, Jeff Candell, Jennie H. Kwon, Hilary Babcock, and  al.

   Infection Control & Hospital Epidemiology, 2024

   Abs HTML

   To improve contact tracing for healthcare workers, we built and configured a Bluetooth low-energy system. We predicted close contacts with great accuracy and provided an additional contact yield of 14.8%. This system would decrease the effective reproduction number by 56% and would unnecessarily quarantine 0.74% of employees weekly.

2023

1. RTSS 2023
   Best Student
   Paper Award

   

   Progressive Neural Compression for Adaptive Image Offloading Under Timing Constraints

   Ruiqi Wang, Hanyang Liu, Jiaming Qiu, Moran Xu, Roch Guérin, and Chenyang Lu

   In 2023 IEEE Real-Time Systems Symposium (RTSS) , 2023

   Awareded Abs arXiv Bib HTML Code

   Best Student Paper Award

   IoT devices are increasingly the source of data for machine learning (ML) applications running on edge servers. Data transmissions from devices to servers are often over local wireless networks whose bandwidth is not just limited but, more importantly, variable. Furthermore, in cyber-physical systems interacting with the physical environment, image offloading is also commonly subject to timing constraints. It is, therefore, important to develop an adaptive approach that maximizes the inference performance of ML applications under timing constraints and the resource constraints of IoT devices. In this paper, we use image classification as our target application and propose progressive neural compression (PNC) as an efficient solution to this problem. Although neural compression has been used to compress images for different ML applications, existing solutions often produce fixed-size outputs that are unsuitable for timing-constrained offloading over variable bandwidth. To address this limitation, we train a multi-objective rateless autoencoder that optimizes for multiple compression rates via stochastic taildrop to create a compression solution that produces features ordered according to their importance to inference performance. Features are then transmitted in that order based on available bandwidth, with classification ultimately performed using the (sub)set of features received by the deadline. We demonstrate the benefits of PNC over state-of-the-art neural compression approaches and traditional compression methods on a testbed comprising an IoT device and an edge server connected over a wireless network with varying bandwidth.

   @inproceedings{10405983,
     author = {Wang, Ruiqi and Liu, Hanyang and Qiu, Jiaming and Xu, Moran and Guérin, Roch and Lu, Chenyang},
     booktitle = {2023 IEEE Real-Time Systems Symposium (RTSS)},
     title = {Progressive Neural Compression for Adaptive Image Offloading Under Timing Constraints},
     year = {2023},
     volume = {},
     number = {},
     pages = {118-130},
     keywords = {Performance evaluation;Image coding;Image edge detection;Bandwidth;Timing;Internet of Things;Servers;neural compression;edge offloading;image classification;real-time transmission},
     doi = {10.1109/RTSS59052.2023.00020},
   }

2. UbiComp 2023

   

   Contact Tracing for Healthcare Workers in an Intensive Care Unit

   Jingwen Zhang, Ruixuan Dai, Ashraf Rjob, Ruiqi Wang, Reshad Hamauon, Jeffrey Candell, Thomas Bailey, Victoria J. Fraser, Maria Cristina Vazquez Guillamet, and Chenyang Lu

   Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Sep 2023

   Abs Bib HTML

   Contact tracing is a powerful tool for mitigating the spread of COVID-19 during the pandemic. Front-line healthcare workers are particularly at high risk of infection in hospital units. This paper presents ContAct TraCing for Hospitals (CATCH), an automated contact tracing system designed specifically for healthcare workers in hospital environments. CATCH employs distributed embedded devices placed throughout a hospital unit to detect close contacts among healthcare workers wearing Bluetooth Low Energy (BLE) beacons. We first identify a set of distinct contact tracing scenarios based on the diverse environmental characteristics of a real-world intensive care unit (ICU) and the different working patterns of healthcare workers in different spaces within the unit. We then develop a suite of novel contact tracing methods tailored for each scenario. CATCH has been deployed and evaluated in the ICU of a major medical center, demonstrating superior accuracy in contact tracing over existing approaches through a wide range of experiments. Furthermore, the real-world case study highlights the effectiveness and efficiency of CATCH compared to standard contact tracing practices.

   @article{10.1145/3610924,
     author = {Zhang, Jingwen and Dai, Ruixuan and Rjob, Ashraf and Wang, Ruiqi and Hamauon, Reshad and Candell, Jeffrey and Bailey, Thomas and Fraser, Victoria J. and Guillamet, Maria Cristina Vazquez and Lu, Chenyang},
     title = {Contact Tracing for Healthcare Workers in an Intensive Care Unit},
     year = {2023},
     issue_date = {September 2023},
     publisher = {Association for Computing Machinery},
     address = {New York, NY, USA},
     volume = {7},
     number = {3},
     url = {https://doi.org/10.1145/3610924},
     doi = {10.1145/3610924},
     journal = {Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.},
     month = sep,
     articleno = {141},
     numpages = {23},
     keywords = {Internet of things (IoT), Bluetooth Low Energy, Contact Tracing, Covid-19},
   }

2022

1. EMSOFT 2022

   

   Adaptive Edge Offloading for Image Classification Under Rate Limit

   Jiaming Qiu, Ruiqi Wang, Ayan Chakrabarti, Roch Guerin, and Chenyang Lu

   IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, Nov 2022

   Abs arXiv Bib HTML Code

   This article considers a setting where embedded devices are used to acquire and classify images. Because of limited computing capacity, embedded devices rely on a parsimonious classification model with uneven accuracy. When local classification is deemed inaccurate, devices can decide to offload the image to an edge server with a more accurate but resource-intensive model. Resource constraints, e.g., network bandwidth, however, require regulating such transmissions to avoid congestion and high latency. This article investigates this offloading problem when transmissions regulation is through a token bucket, a mechanism commonly used for such purposes. The goal is to devise a lightweight, online offloading policy that optimizes an application-specific metric (e.g., classification accuracy) under the constraints of the token bucket. This article develops a policy based on a deep Q-network (DQN), and demonstrates both its efficacy and the feasibility of its deployment on embedded devices. Of note is the fact that the policy can handle complex input patterns, including correlation in image arrivals and classification accuracy. The evaluation is carried out by performing image classification over a local testbed using synthetic traces generated from the ImageNet image classification benchmark.

   @article{article,
     author = {Qiu, Jiaming and Wang, Ruiqi and Chakrabarti, Ayan and Guerin, Roch and Lu, Chenyang},
     year = {2022},
     month = nov,
     pages = {1-1},
     title = {Adaptive Edge Offloading for Image Classification Under Rate Limit},
     volume = {41},
     journal = {IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems},
     doi = {10.1109/TCAD.2022.3197533},
   }