ABSTRACT
Pedestrian detection is one of the challenging tasks in the technology of autonomous driving. Recently, the object detection network of you only look once (YOLO), especially YOLOv3 and YOLOv3-tiny have demonstrated a high level of pedestrian detection performance on a powerful GPU card such as Pascal Titan X. However, it is still challenging to use YOLOv3 and YOLOv3-tiny on embedded GPU system due to their large network size. In this paper, we present a lightweight YOLOv3-mobile network by refining the architecture of YOLOv3-tiny to improve its pedestrian detection efficiency on embedded GPUs such as Nvidia Jetson TX1. The experimental results showed that the proposed framework can accelerate the frame rate per second (FPS) from 18 FPS to 37 FPS with comparable mean average precision (mAP).
- Sha Ding, Fei Long, Huijin Fan, Lei Liu, and Yongji Wang. 2019. A novel YOLOv3-tiny network for unmanned airship obstacle detection. In 2019 IEEE 8th Data Driven Control and Learning Systems Conference (DDCLS). IEEE, 277–281.Google ScholarCross Ref
- Andreas Ess, Bastian Leibe, Konrad Schindler, and Luc Van Gool. 2008. A mobile vision system for robust multi-person tracking. In 2008 IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 1–8.Google ScholarCross Ref
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition. 770–778.Google ScholarCross Ref
- Rasheed Hussain and Sherali Zeadally. 2018. Autonomous cars: Research results, issues, and future challenges. IEEE Communications Surveys & Tutorials 21, 2 (2018), 1275–1313.Google ScholarCross Ref
- Forrest N Iandola, Matthew W Moskewicz, Khalid Ashraf, and Kurt Keutzer. 2016. Firecaffe: near-linear acceleration of deep neural network training on compute clusters. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2592–2600.Google ScholarCross Ref
- SeonTaek Oh, Ji-Hwan You, and Young-Keun Kim. 2020. Implementation of Compressed YOLOv3-tiny on FPGA-SoC. In 2020 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia). IEEE, 1–4.Google ScholarCross Ref
- Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi. 2016. You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition. 779–788.Google ScholarCross Ref
- Joseph Redmon and Ali Farhadi. 2017. YOLO9000: better, faster, stronger. In Proceedings of the IEEE conference on computer vision and pattern recognition. 7263–7271.Google ScholarCross Ref
- Joseph Redmon and Ali Farhadi. 2018. Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767(2018).Google Scholar
- Eduard Zadobrischi and Mihai Negru. 2020. Pedestrian detection based on TensorFlow YOLOv3 embedded in a portable system adaptable to vehicles. In 2020 International Conference on Development and Application Systems (DAS). IEEE, 21–26.Google ScholarCross Ref
Recommendations
TX-RSA: A High Performance RSA Implementation Scheme on NVIDIA Tegra X2
Wireless Algorithms, Systems, and ApplicationsAbstractDriven by computer vision and autopilot industries, embedded graphics processing units (GPUs) are now rapidly achieving extraordinary computing power, such NVIDIA Tegra K1/X1/X2, which are widely used in embedded environments such as mobile phones,...
Computing prestack Kirchhoff time migration on general purpose GPU
This paper introduces how to optimize a practical prestack Kirchhoff time migration program by the Compute Unified Device Architecture (CUDA) on a general purpose GPU (GPGPU). A few useful optimization methods on GPGPU are demonstrated, such as how to ...
Accelerating the discontinuous Galerkin method for seismic wave propagation simulations using the graphic processing unit (GPU)-single-GPU implementation
We have successfully ported an arbitrary high-order discontinuous Galerkin (ADER-DG) method for solving the three-dimensional elastic seismic wave equation on unstructured tetrahedral meshes to an Nvidia Tesla C2075 GPU using the Nvidia CUDA programming ...
Comments