Shaowu Chen

TinyDrop: Tiny Model Guided Token Dropping for Vision Transformers

Guoxin Wang, Qingyuan Wang, Shaowu Chen, Deepu John

Vision Transformers (ViTs) achieve strong performance in image classification but incur high computational costs from processing all image tokens. To reduce inference costs in large ViTs without compromising accuracy, we propose Tiny-Drop, a training-free token dropping framework guided by a lightweight vision …

ie, cn (code pays fourni par la source)

CF-Net: A Cross-Feature Reconstruction Network for High-Accuracy 1-Bit Target Classification

Jundong Qi, Weize Sun, Shaowu Chen, Lei Huang et autres

cn, jp (code pays fourni par la source)

MoCrop: Training Free Motion Guided Cropping for Efficient Video Action Recognition

Binhua Huang, Wendong Yao, Shaowu Chen, Guoxin Wang et autres

Standard video action recognition models often process typically resized full frames, suffering from spatial redundancy and high computational costs. To address this, we introduce MoCrop, a motion-aware adaptive cropping module designed for efficient video action recognition in the compressed domain. Leveraging Motion …

Lightweight Target Detection Network for Security Radar and FPGA Acceleration

Weize Sun, Yonglong Zhang, Shaowu Chen, Lei Huang et autres

Radar target detection technology has been widely used in many fields. In some application scenarios, such as security screening at a high-traffic airport, we need a flexible and real-time contraband detection system to ensure the safety of passengers. However, the object detection …

cn (code pays fourni par la source)

POCKET: Pruning random convolution kernels for time series classification from a feature selection perspective

Shaowu Chen, Weize Sun, Xiao Peng Li, Qingyuan Wang et autres

ie, cn (code pays fourni par la source)

POCKET: Pruning Random Convolution Kernels for Time Series Classification from a Feature Selection Perspective

Shaowu Chen, Weize Sun, Lei Huang, Xiao Peng Li et autres

In recent years, two competitive time series classification models, namely, ROCKET and MINIROCKET, have garnered considerable attention due to their low training cost and high accuracy. However, they rely on a large number of random 1-D convolutional kernels to comprehensively capture features, …

ie, cn (code pays fourni par la source)

Mechanisms of ferroptosis in Alzheimer's disease and therapeutic effects of natural plant products: A review

Zhao Da, Kailin Yang, Hua Claria Guo, Jinsong Zeng et autres

Neurodegenerative diseases, such as Alzheimer's disease (AD), are characterized by massive loss of specific neurons. It is a progressive disabling, severe and fatal complex disease. Due to its complex pathogenesis and limitations of clinical treatment strategies, it poses a serious medical challenge …

cn (code pays fourni par la source)

WHC: Weighted Hybrid Criterion for Filter Pruning on Convolutional Neural Networks

Shaowu Chen, Weize Sun, Lei Huang

Filter pruning has attracted increasing attention in recent years for its capacity in compressing and accelerating convolutional neural networks. Various data-independent criteria, including norm-based and relationship-based ones, were proposed to prune the most unimportant filters. However, these state-of-the-art criteria fail to fully …

cn (code pays fourni par la source)

[Effect and safety of double plasma molecular absorption in treatment of patients with severe wasp stings injury: a multicenter historical cohort study].

Tingqiu Wei, Huafeng Liao, Junxin Xiao, Dezhi Zeng et autres

OBJECTIVE: To compare the effect and safety of continuous veno-venous hemofiltration (CVVH)+double plasma molecular absorption (DPMA)+hemoperfusion (HP), CVVH+HP, and CVVH+plasma exchange (PE) in treatment of patient with severe wasp stings injury. METHODS: Multicenter, historical cohort study and superiority test were used. From …

cn (code pays fourni par la source)

WHC: Weighted Hybrid Criterion for Filter Pruning on Convolutional Neural Networks

Shaowu Chen, Weize Sun, Lei Huang

Filter pruning has attracted increasing attention in recent years for its capacity in compressing and accelerating convolutional neural networks. Various data-independent criteria, including norm-based and relationship-based ones, were proposed to prune the most unimportant filters. However, these state-of-the-art criteria fail to fully …

Radar Target Recognition with Self-Supplementary Neural Network

Xuhua Huang, Weize Sun, Lei Huang, Shaowu Chen et autres

Radar target recognition is an important problem in radar system and had been widely studied in the pass decades. In this paper, we first build a radar target recognition system that classifies the target based on$i$successive frames of echo values. A baseline …

cn (code pays fourni par la source)

Deep Learning-Based Non-synchronous Measurement for Broadband Sound Source Localization

Guitong Chen, Long Chen, Weize Sun, Shaowu Chen et autres

Array-based acoustic beamforming methods are prevalently used in many practical fields. However, limited by the size of the array and sensor elements, these methods cannot achieve the best performance. To mitigate the limitations, Non-synchronous Measurement (NSM), which synthesizes a virtual array with …

cn (code pays fourni par la source)