http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Conditional Generative Adversarial Network를 이용한 이미지 방사 왜곡 보정 영상 생성
박동훈,Vijay Kakani,김학일 제어·로봇·시스템학회 2019 제어·로봇·시스템학회 논문지 Vol.25 No.11
This article describes a method for radial un-distortion of image using a conditional generative adversarial network. The proposed network consists of a generator which has a similar shape of U-Net and a shallow discriminator. The proposed model is trained by using perceptual loss, content loss and adversarial loss over the PASCAL VOC datasets where each sample image is distorted by one-parameter radial distortion model and inserted as a condition. The experimental results are compared with traditional radial un-distortion models such as Bukhari’s and Rong’s methods, and demonstrate not only 12-times faster distortion correction speeds but also a significant improvement in PSNR and SSIM. Additionally, the corrected images show an improved performance in object detection.
Edge Device Deployment of Multi-Tasking Network self-Driving Operations
Shokhrukh Miraliev,Shakhboz Abdigapporov,Jumabek Alikhanov,Vijay Kakani,Hakil Kim 한국차세대컴퓨팅학회 2022 한국차세대컴퓨팅학회 학술대회 Vol.2022 No.10
A safe and robust autonomous driving system relies on accurate perception of the environment for application-oriented scenarios. This paper proposes deployment of the three most crucial tasks (i.e., object detection, drivable area segmentation and lane detection tasks) on embedded system for self-driving operations. To achieve this research objective, multi-tasking network is utilized with a simple encoder-decoder architecture. Comprehensive and extensive comparisons for two models based on different backbone networks are performed. All training experiments are performed on server while Nvidia Jetson Xavier NX is chosen as deployment device.
Performance Comparison of Backbone Networks for Multi-Tasking in Self-Driving Operations
Shakhboz Abdigapporov,Shokhrukh Miraliev,Jumabek Alikhanov,Vijay Kakani,Hakil Kim 제어로봇시스템학회 2022 제어로봇시스템학회 국제학술대회 논문집 Vol.2022 No.11
In the era of big data, increased focus has been on improving neural network based Deep Learning models. This led to various classification networks which can be used as a backbone in multi-task learning. However, depending on the selected backbone, multi-tasking performance differs. While given backbone network shows better performance on a detection task, does not mean such performance generalizes in segmentation task as well. Detailed investigations should be conducted to achieve best inference speed-accuracy trade-off prior to implementing a single neural network, which handles multiple tasks. In this research, the performance comparison among EfficientNet, ResNet101, VGG16, ResNet50 and MobilenetV2 on the Berkeley Driving Dataset (BDD100K) for autonomous driving using multi-tasking architecture are provided. Backbones that offer best time-accuracy trade-off for multi-task learning are evaluated. Implemented architecture contains three most crucial tasks in self-driving operations, object detection, drivable area segmentation and lane detection. EfficientNet based model showed the best mAP on the object detection task, as well as on the segmentation tasks, extracting both the long and wide roads with accurate lane lines. The model with MobilenetV2 backbone however, demonstrates the fastest inference speed with relatively lower performance in all tasks.