mlsanityData to 14 Oct 2026

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

arXiv 2108.00392 ↗First posted 1 Aug 2021cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Latest CNN-based object detection models are quite accurate but require a high-performance GPU to run in real-time. They still are heavy in terms of memory size and speed for an embedded system with limited memory space. Since the object detection for autonomous system is run on an embedded processor, it is preferable to compress the detection network as light as possible while preserving the detection accuracy. There are several popular lightweight detection models but their accuracy is too low for safe driving applications. Therefore, this paper proposes a new object detection model, referred as YOffleNet, which is compressed at a high ratio while minimizing the accuracy loss for real-time and safe driving application on an autonomous system. The backbone network architecture is based on YOLOv4, but we could compress the network greatly by replacing the high-calculation-load CSP DenseNet with the lighter modules of ShuffleNet. Experiments with KITTI dataset showed that the proposed YOffleNet is compressed by 4.7 times than the YOLOv4-s that could achieve 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 to 85.8% mAP, that is only 2.6% lower than YOLOv4-s. Thus, the proposed network showed a high potential to be deployed on the embedded system of the autonomous system for the real-time and accurate object detection applications.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 24 of its 29 references that are in the map. First posted on arXiv 1 Aug 2021.

Builds on

24 of 29 references.

References · 24

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  3. 2017Mask R-CNNcited by 7,277
  4. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
  5. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  6. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
  7. 2015SSD: Single Shot MultiBox Detectorcited by 3,130
  8. 2018YOLOv3: An Incremental Improvementcited by 2,659
  9. 2019Searching for MobileNetV3cited by 1,994
  10. 2020YOLOv4: Optimal Speed and Accuracy of Object Detectioncited by 1,630
  11. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  12. 2016YOLO9000: Better, Faster, Strongercited by 1,431
12 more
  1. 2017ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devicescited by 1,200
  2. 2018ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Designcited by 1,136
  3. 2016SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model sizecited by 1,060
  4. 2018Path Aggregation Network for Instance Segmentationcited by 798
  5. 2014Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognitioncited by 720
  6. 2019CSPNet: A New Backbone that can Enhance Learning Capability of CNNcited by 381
  7. 2017A Survey of Model Compression and Acceleration for Deep Neural Networkscited by 301
  8. 2017DSSD : Deconvolutional Single Shot Detectorcited by 131
  9. 2017Single-Shot Refinement Neural Network for Object Detectioncited by 129
  10. 2018Robust Deep Reinforcement Learning for Security and Safety in Autonomous Vehicle Systemscited by 17
  11. 2018DEEPEYE: A Compact and Accurate Video Comprehension at Terminal Devices Compressed with Quantization and Tensorizationcited by 1
  12. 2020FRDet: Balanced and Lightweight Object Detector based on Fire-Residual Modules for Embedded Processor of Autonomous Drivingcited by 1

Built on by

Not cited yet.