TY - GEN
T1 - Camera-wise Training for Enhanced Omni-directional 2D Object Detection
AU - Jeon, Hyung Joon
AU - Tran, Duong Nguyen Ngoc
AU - Pham, Long Hoang
AU - Nguyen, Huy Hung
AU - Tran, Tai Huu Phuong
AU - Jeon, Jae Wook
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - In this paper, we propose a method to perform training and inference with multiple instances of the same deep neural network architecture on images taken from cameras of different directions. Across multiple cameras, depending on each of their directional characteristics, objects viewed from the cameras can form slightly different distributions in visual features. Regarding this, we emphasize the importance of camera-wise training on multiple instances of a given deep neural network for object detection. Given the Waymo Open Perception Dataset, we used multiple instances of the YOLOv5x6 architecture and trained each of them per camera. Such a training scheme on the Training Set achieves better training progression, and the inference results are shown to have AP/L1 as high as 0.6679 on the Testing Set.
AB - In this paper, we propose a method to perform training and inference with multiple instances of the same deep neural network architecture on images taken from cameras of different directions. Across multiple cameras, depending on each of their directional characteristics, objects viewed from the cameras can form slightly different distributions in visual features. Regarding this, we emphasize the importance of camera-wise training on multiple instances of a given deep neural network for object detection. Given the Waymo Open Perception Dataset, we used multiple instances of the YOLOv5x6 architecture and trained each of them per camera. Such a training scheme on the Training Set achieves better training progression, and the inference results are shown to have AP/L1 as high as 0.6679 on the Testing Set.
KW - autonomous driving
KW - camera-wise training
KW - object detection
KW - YOLOv5x6
UR - https://www.scopus.com/pages/publications/85143891462
U2 - 10.1109/IECON49645.2022.9968791
DO - 10.1109/IECON49645.2022.9968791
M3 - Conference contribution
AN - SCOPUS:85143891462
T3 - IECON Proceedings (Industrial Electronics Conference)
BT - IECON 2022 - 48th Annual Conference of the IEEE Industrial Electronics Society
PB - IEEE Computer Society
T2 - 48th Annual Conference of the IEEE Industrial Electronics Society, IECON 2022
Y2 - 17 October 2022 through 20 October 2022
ER -