Benchmarking Jetson Edge Devices with an End-to-end Video-based Anomaly Detection System
Authors:
Hoang Viet Pham,
Thinh Gia Tran,
Chuong Dinh Le,
An Dinh Le,
Hien Bich Vo
Abstract:
Innovative enhancement in embedded system platforms, specifically hardware accelerations, significantly influence the application of deep learning in real-world scenarios. These innovations translate human labor efforts into automated intelligent systems employed in various areas such as autonomous driving, robotics, Internet-of-Things (IoT), and numerous other impactful applications. NVIDIA's Jet…
▽ More
Innovative enhancement in embedded system platforms, specifically hardware accelerations, significantly influence the application of deep learning in real-world scenarios. These innovations translate human labor efforts into automated intelligent systems employed in various areas such as autonomous driving, robotics, Internet-of-Things (IoT), and numerous other impactful applications. NVIDIA's Jetson platform is one of the pioneers in offering optimal performance regarding energy efficiency and throughput in the execution of deep learning algorithms. Previously, most benchmarking analysis was based on 2D images with a single deep learning model for each comparison result. In this paper, we implement an end-to-end video-based crime-scene anomaly detection system inputting from surveillance videos and the system is deployed and completely operates on multiple Jetson edge devices (Nano, AGX Xavier, Orin Nano). The comparison analysis includes the integration of Torch-TensorRT as a software developer kit from NVIDIA for the model performance optimisation. The system is built based on the PySlowfast open-source project from Facebook as the coding template. The end-to-end system process comprises the videos from camera, data preprocessing pipeline, feature extractor and the anomaly detection. We provide the experience of an AI-based system deployment on various Jetson Edge devices with Docker technology. Regarding anomaly detectors, a weakly supervised video-based deep learning model called Robust Temporal Feature Magnitude Learning (RTFM) is applied in the system. The approach system reaches 47.56 frames per second (FPS) inference speed on a Jetson edge device with only 3.11 GB RAM usage total. We also discover the promising Jetson device that the AI system achieves 15% better performance than the previous version of Jetson devices while consuming 50% less energy power.
△ Less
Submitted 12 September, 2023; v1 submitted 28 July, 2023;
originally announced July 2023.
Sheaf-theoretic self-filtering network of low-cost sensors for local air quality monitoring: A causal approach
Authors:
Anh-Duy Pham,
Chuong Dinh Le,
Hoang Viet Pham,
Thinh Gia Tran,
Dat Thanh Vo,
Chau Long Tran,
An Dinh Le,
Hien Bich Vo
Abstract:
Sheaf theory, which is a complex but powerful tool supported by topological theory, offers more flexibility and precision than traditional graph theory when it comes to modeling relationships between multiple features. In the realm of air quality monitoring, this can be incredibly useful in detecting sudden changes in local dust particle density, which can be difficult to accurately measure using…
▽ More
Sheaf theory, which is a complex but powerful tool supported by topological theory, offers more flexibility and precision than traditional graph theory when it comes to modeling relationships between multiple features. In the realm of air quality monitoring, this can be incredibly useful in detecting sudden changes in local dust particle density, which can be difficult to accurately measure using commercial instruments. Traditional methods for air quality measurement often rely on calibrating the measurement with public standard instruments or calculating the measurements moving average over a constant period. However, this can lead to an incorrect index at the measurement location, as well as an oversmoothing effect on the signal. In this study, we propose a compact device that uses sheaf theory to detect and count vehicles as a local air quality change-causing factor. By inferring the number of vehicles into the PM2.5 index and propagating it into the recorded PM2.5 index from low-cost air monitoring sensors such as PMS7003 and BME280, we can achieve self-correction in real-time. Plus, the sheaf-theoretic method allows for easy scaling to multiple nodes for further filtering effects. By implementing sheaf theory in air quality monitoring, we can overcome the limitations of traditional methods and provide more accurate and reliable results.
△ Less
Submitted 29 December, 2022;
originally announced December 2022.
