Butterfly network: a convolutional neural network with a new architecture for multi-scale semantic segmentation of pedestrians
Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Standard
Butterfly network : a convolutional neural network with a new architecture for multi-scale semantic segmentation of pedestrians. / Alavianmehr, M.A.; Helfroush, M.S.; Danyali, H.; Tashk, Ashkan.
I: Journal of Real-Time Image Processing, Bind 20, 9, 2023.Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Butterfly network
T2 - a convolutional neural network with a new architecture for multi-scale semantic segmentation of pedestrians
AU - Alavianmehr, M.A.
AU - Helfroush, M.S.
AU - Danyali, H.
AU - Tashk, Ashkan
PY - 2023
Y1 - 2023
N2 - The detection of multi-scale pedestrians is one of the challenging tasks in pedestrian detection applications. Moreover, the task of small-scale pedestrian detection, i.e., accurate localization of pedestrians as low-scale target objects, can help solve the issue of occluded pedestrian detection as well. In this paper, we present a fully convolutional neural network with a new architecture and an innovative, fully detailed supervision for semantic segmentation of pedestrians. The proposed network has been named butterfly network (BF-Net) because of its architecture analogous to a butterfly. The proposed BF-Net preserves the ability of simplicity so that it can process static images with a real-time image processing rate. The sub-path blocks embedded in the architecture of the proposed BF-Net provides a higher accuracy for detecting multi-scale objective targets including the small ones. The other advantage of the proposed architecture is replacing common batch normalization with conditional one. In conclusion, the experimental results of the proposed method demonstrate that the proposed network outperform the other state-of-the-art networks such as U-Net + + , U-Net3 + , Mask-RCNN, and Deeplabv3 + for the semantic segmentation of the pedestrians.
AB - The detection of multi-scale pedestrians is one of the challenging tasks in pedestrian detection applications. Moreover, the task of small-scale pedestrian detection, i.e., accurate localization of pedestrians as low-scale target objects, can help solve the issue of occluded pedestrian detection as well. In this paper, we present a fully convolutional neural network with a new architecture and an innovative, fully detailed supervision for semantic segmentation of pedestrians. The proposed network has been named butterfly network (BF-Net) because of its architecture analogous to a butterfly. The proposed BF-Net preserves the ability of simplicity so that it can process static images with a real-time image processing rate. The sub-path blocks embedded in the architecture of the proposed BF-Net provides a higher accuracy for detecting multi-scale objective targets including the small ones. The other advantage of the proposed architecture is replacing common batch normalization with conditional one. In conclusion, the experimental results of the proposed method demonstrate that the proposed network outperform the other state-of-the-art networks such as U-Net + + , U-Net3 + , Mask-RCNN, and Deeplabv3 + for the semantic segmentation of the pedestrians.
U2 - 10.1007/s11554-023-01273-z
DO - 10.1007/s11554-023-01273-z
M3 - Journal article
C2 - 36748032
VL - 20
JO - Journal of Real-Time Image Processing
JF - Journal of Real-Time Image Processing
SN - 1861-8200
M1 - 9
ER -
ID: 342053420