Cloud Versus Edge Deployment Strategies of Real-Time Face Recognition Inference
Abstract
Choosing the appropriate deployment strategy for any Deep Learning (DL) project in a production environment has always been the most challenging problem for industrial practitioners. There are several conflicting constraints and controversial approaches when it comes to deployment. Among these problems, the deployment on cloud versus the deployment on edge represents a common dilemma. In a nutshell, each approach provides benefits where the other would have limitations. This paper presents a real-world case study on deploying a face recognition application using MTCNN detector and FaceNet recognizer. We report the challenges faced to decide on the best deployment strategy. We propose three inference architectures for the deployment, including cloud-based, edge-based, and hybrid. Furthermore, we evaluate the performance of face recognition inference on different cloud-based and edge-based GPU platforms. We consider different models of Jetson boards for the edge (Nano, TX2, Xavier NX, Xavier AGX) and various GPUs for the cloud (GTX 1080, RTX 2080Ti, RTX 2070, and RTX 8000). We also investigate the effect of deep learning model optimization using TensorRT and TFLite compared to a standard Tensorflow GPU model, and the effect of input resolution. We provide a benchmarking study for all these devices in terms of frames per second, execution times, energy and memory usages. After conducting a total of 294 experiments, the results demonstrate that the TensorRT optimization provides the fastest execution on all cloud and edge devices, at the expense of significantly larger energy consumption (up to +40% and +35% for edge and cloud devices, respectively, compared to Tensorflow). Whereas TFLite is the most efficient framework in terms of memory and power consumption, while providing significantly less (-4% to -62%) processing acceleration than TensorRT.Practitioners Note:The study reported in this paper presents the real-challenges that we faced during our development and deployment of a face-recognition application both on the edge and on the cloud, and the solutions we have developed to solve these problems. The code, results, and interactive analytic dashboards of this paper will be put public upon publication.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
YOLO's approach diverged from traditional two-stage object detection architectures by proposing a unified architecture that predicts bounding boxes and class probabilities concurrently within real-time processing constraints.
“YOLOs innovative approach diverged from the traditional two-stage object detection architectures, proposing a unified architecture with the ability to predict bounding boxes and class probabilities concurrently, all within the exigencies of real-time processing [8].”
From YOLOv1 to v8: Unveiling Each Variant–A Comprehensive Review of YOLO (2024) · cites this paper as [8]
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 56,035 · see citing works
YOLOv4: Optimal Speed and Accuracy of Object Detection
Bochkovskiy, A., Wang, C.-Y., & Liao, H.-Y. M. (2020) · arXiv · doi:10.48550/arxiv.2004.10934 · Cited by 10,307 · see citing works
WaveNet: a generative model for raw audio
Oord, A. V. D., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., Kalchbrenner, N., Senior, A., & Kavukcuoglu, K. (2016) · arXiv · doi:10.48550/arxiv.1609.03499 · Cited by 4,911 · see citing works
VGGFace2: A Dataset for Recognising Faces across Pose and Age
Cao, Q., Shen, L., Xie, W., Parkhi, O. M., & Zisserman, A. (2018) · 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018) · doi:10.1109/fg.2018.00020 · Cited by 3,032 · see citing works
Deep Learning With Edge Computing: A Review
Chen, J., & Ran, X. (2019) · Proceedings of the IEEE · doi:10.1109/jproc.2019.2921977 · Cited by 1,540 · cited 1× · see citing works
“…and did not investigate the effect of model optimization using TFLite and TensorRT, as we proposed in this paper Wang et al. [14], and Chen and Ran [25] presented theoretical surveys that compared several edge computation and communication modes used in the literature (integral or partial…”
MethodConvergence of Edge Computing and Deep Learning: A Comprehensive Survey
Wang, X., Han, Y., Leung, V. C. M., Niyato, D., Yan, X., & Chen, X. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.2970550 · Cited by 1,498 · cited 2× · see citing works
“In recent years, edge computing emerged as an increasing trend compared to cloud computing for several reasons [14].”
BackgroundMethodEdge Assisted Real-time Object Detection for Mobile Augmented Reality
Liu, L., Li, H., & Gruteser, M. (2019) · The 25th Annual International Conference on Mobile Computing and Networking · doi:10.1145/3300061.3300116 · Cited by 552 · see citing works
Deep Learning Empowered Task Offloading for Mobile Edge Computing in Urban Informatics
Zhang, K., Zhu, Y., Leng, S., He, Y., Maharjan, S., & Zhang, Y. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2019.2903191 · Cited by 309 · see citing works
Towards Fair and Privacy-Preserving Federated Deep Models
Lyu, L., Yu, J., Nandakumar, K., Li, Y., Ma, X., Jin, J., Yu, H., & Ng, K. S. (2020) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2020.2996273 · Cited by 250 · cited 1× · see citing works
“Several recent research works attempted to develop privacy-preserving deep learning models [62], [63], [64], [65].”
BackgroundQuantifying the Impact of Edge Computing on Mobile Applications
Hu, W., Gao, Y., Ha, K., Wang, J., Amos, B., Chen, Z., Pillai, P., & Satyanarayanan, M. (2016) · Proceedings of the 7th ACM SIGOPS Asia-Pacific Workshop on Systems · doi:10.1145/2967360.2967369 · Cited by 171 · see citing works
EdgeEye: An Edge Service Framework for Real-time Intelligent Video Analytics
Liu, P., Qi, B., & Banerjee, S. (2018) · Proceedings of the 1st International Workshop on Edge Systems, Analytics and Networking · doi:10.1145/3213344.3213345 · Cited by 93 · see citing works
Adaptive deep learning model selection on embedded systems
Taylor, B., Marco, V. S., Wolff, W., Elkhatib, Y., & Wang, Z. (2018) · Proceedings of the 19th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems · doi:10.1145/3211332.3211336 · Cited by 90 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
YOLOv1 to v8: Unveiling Each Variant–A Comprehensive Review of YOLO
Hussain, M. (2024) · IEEE Access · doi:10.1109/access.2024.3378568 · Cited by 492 · see citing works
A Review of Embedded Machine Learning Based on Hardware, Application, and Sensing Scheme
Biglari, A., & Tang, W. (2023) · Sensors · doi:10.3390/s23042131 · Cited by 86 · see citing works
- Authors
- Koubaa, A., Ammar, A., Kanhouch, A., & AlHabashi, Y.
- Affiliations
- Prince Sultan University (SA)
- Venue
- IEEE Transactions on Network Science and Engineering
- Published
- 2022-01-01
- Volume
- 9
- Issue
- 1
- Pages
- 143-160
- ISSN
- 2327-4697 (ISSN-L)
- 2334-329X
- OpenAlex
- W3128702906
- Cited by
- 69
- References
- 13
- Citations by year
- 2026: 9 · 2025: 23 · 2024: 10 · 2023: 13 · 2022: 9 · 2021: 5
- FWCI
- 3.33
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Network Science and Engineering (publishedVersion)
- Funders
- Prince Sultan University (SEED-2020-05, SEED-CCIS-2020-05)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Keywords
- face recognition, real-time inference, cloud deployment, edge deployment, hybrid deployment, GPU benchmarking, TensorRT, TensorFlow Lite, model optimization, MTCNN, FaceNet, energy consumption, memory usage, input resolution, deployment strategies
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
