Skip to Main content Skip to Navigation
New interface
Conference papers

Optimization of Deep-Learning Detection of Humans in Marine Environment on Edge Devices

Mostafa Rizk 1, 2, 3 Dominique Heller 4 Ronan Douguet 4 Amer Baghdadi 1, 3 Jean-Philippe Diguet 5 
1 Lab-STICC_2AI - Equipe Algorithm Architecture Interactions
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance : UMR6285
4 Lab-STICC_SHAKER - Equipe Software/HArdware and unKnown Environment inteRactions
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance : UMR6285
Abstract : Artificial intelligence (AI) detection techniques based on convolution neural networks (CNNs) require high computations and memory. Their deployment on embedded edge devices, with reduced resources and power budget, is highly hindered especially for applications that requires real-time inference. Several optimization methods such as pruning, quantization and using shallow networks, are mainly utilized to overcome this limitation but at the cost of degradation in detection performance. However, efficient approaches for training and inference have been recently introduced to lower such degradation. This work investigates the use of these approaches to optimize the popular You Only Look Once (YOLO) network targeting various emerging edge devices (Nvidia Jetson Xavier AGX, AMD-Xilinx Kria KV260 Vision AI Kit, and Movidius Myriad X VPU) in order to enhance the detection of humans in maritime environment.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03789216
Contributor : Mostafa RIZK Connect in order to contact the contributor
Submitted on : Tuesday, September 27, 2022 - 11:58:06 AM
Last modification on : Saturday, October 1, 2022 - 3:50:53 AM

File

ICECS2022_Final16092022.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03789216, version 1

Citation

Mostafa Rizk, Dominique Heller, Ronan Douguet, Amer Baghdadi, Jean-Philippe Diguet. Optimization of Deep-Learning Detection of Humans in Marine Environment on Edge Devices. ICECS 2022: IEEE International Conference on Electronics Circuits and Systems, Oct 2022, Glasgow, United Kingdom. ⟨hal-03789216⟩

Share

Metrics

Record views

36

Files downloads

10