English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Latest CNN-based object detection models are quite accurate but require ahigh-performance GPU to run in real-time. They still are heavy in terms ofmemory size and speed for an embedded system with limited memory space. Sincethe object detection for autonomous system is run on an embedded processor, itis preferable to compress the detection network as light as possible whilepreserving the detection accuracy. There are several popular lightweightdetection models but their accuracy is too low for safe driving applications.Therefore, this paper proposes a new object detection model, referred asYOffleNet, which is compressed at a high ratio while minimizing the accuracyloss for real-time and safe driving application on an autonomous system. Thebackbone network architecture is based on YOLOv4, but we could compress thenetwork greatly by replacing the high-calculation-load CSP DenseNet with thelighter modules of ShuffleNet. Experiments with KITTI dataset showed that theproposed YOffleNet is compressed by 4.7 times than the YOLOv4-s that couldachieve as fast as 46 FPS on an embedded GPU system(NVIDIA Jetson AGX Xavier).Compared to the high compression ratio, the accuracy is reduced slightly to85.8% mAP, that is only 2.6% lower than YOLOv4-s. Thus, the proposed networkshowed a high potential to be deployed on the embedded system of the autonomoussystem for the real-time and accurate object detection applications.

Developing a Compressed Object Detection Model based on YOLOv4 for Deployment on Embedded GPU Platform of Autonomous System