Ambiguity-aware robust teacher (ART): Enhanced self-knowledge distillation framework with pruned teacher network

Y Cho, G Ham, JH Lee, D Kim - Pattern Recognition, 2023 - Elsevier
Self-knowledge distillation (self-KD) methods, which use a student model itself as the
teacher model instead of a large and complex teacher model, are currently a subject of …

An ensemble learning aided computer vision method with advanced color enhancement for corroded bolt detection in tunnels

L Tan, T Tang, D Yuan - Sensors, 2022 - mdpi.com
Bolts, as the basic units of tunnel linings, are crucial to safe tunnel service. Caused by the
moist and complex environment in the tunnel, corrosion becomes a significant defect of …

TFDet: Target-aware fusion for RGB-T pedestrian detection

X Zhang, X Zhang, J Wang, J Ying… - … on Neural Networks …, 2024 - ieeexplore.ieee.org
Pedestrian detection plays a critical role in computer vision as it contributes to ensuring
traffic safety. Existing methods that rely solely on RGB images suffer from performance …

基于改进YOLOX 的遥感影像目标检测算法

李美霖, 芮杰, 金飞, 刘智, 林雨准 - 吉林大学学报(地球科学版), 2023 - xuebao.jlu.edu.cn
目标检测是遥感影像处理中一项基础性和常规性的工作. 本文基于YOLOX (you only look once
X) 网络进行改进, 设计了一种针对遥感影像目标的检测算法. 首先在特征提取模块PANet (path …

Region-aware RGB and near-infrared image fusion

J Ying, C Tong, Z Sheng, B Yao, SY Cao, H Yu… - Pattern Recognition, 2023 - Elsevier
This paper proposes a region-aware fusion method, called RaIF, for RGB and near-infrared
(NIR) outdoor scenery image fusion. The method is motivated by the observation that current …

Residual Feature-Reutilization Inception Network

Y He, W Song, L Li, T Zhan, W Jiao - Pattern Recognition, 2024 - Elsevier
Capturing feature information effectively is of great importance in the field of computer vision.
With the development of convolutional neural networks, concepts like residual connection …

[HTML][HTML] Deep Recyclable Trash Sorting Using Integrated Parallel Attention

H Lin, X Zhang, J Yu, J Xiang, HL Shen - Sensors, 2024 - mdpi.com
Sorting recyclable trash is critical to reducing energy consumption and mitigating
environmental pollution. Currently, trash sorting heavily relies on manpower. Computer …

Residual Feature-Reutilization Inception Network for Image Classification

Y He, W Song, L Li, T Zhan, W Jiao - arXiv preprint arXiv:2412.19433, 2024 - arxiv.org
Capturing feature information effectively is of great importance in the field of computer vision.
With the development of convolutional neural networks (CNNs), concepts like residual …

Data Efficient Acoustic Scene Classification using Teacher-Informed Confusing Class Instruction

JJS Yeo, EL Tan, J Bai, S Peksi, WS Gan - arXiv preprint arXiv:2409.11964, 2024 - arxiv.org
In this technical report, we describe the SNTL-NTU team's submission for Task 1 Data-
Efficient Low-Complexity Acoustic Scene Classification of the detection and classification of …

DAT-SegNet: a medical multi-organ segmentation network based on deformation attention and transformer

W Huang, C Liu, R Li, P Chen - Third International Conference …, 2023 - spiedigitallibrary.org
With the development of Transformer and its derivatives over the last two years, several
research have integrated Transformer with CNN or in instead of CNN to advance medical …