Author name cluster

Xuan Nie

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers

1 author row

EAAI Journal 2025 Journal Article

An integrated multi-scale context-aware network for efficient desnowing

Samuel Akwasi Agyemang
Haobin Shi
Xuan Nie
Nana Yaw Asabere

Adverse snow conditions, significantly degrade visual data quality, resulting in substantial performance drops in computer vision systems. We propose an advanced neural network architecture designed for desnowing tasks, leveraging multi-scale features and context-aware attention mechanisms. Our novel network integrates a series of enhancements including ghost encoders for efficient feature representation and extraction through the utilization of linear transformations, and the multi-scale transformer bottleneck which combines the atrous spatial pyramid pooling with a vision transformer in the bottleneck to strengthen multi-scale feature extraction and captures local and global dependencies. Additionally, the SimGC attention module combines the simple parameter-free attention and global context attention, to enhance local and global spatial features. Lastly, the feature enhancement module is used to refine the final reconstruction. Our experimental results demonstrate that the proposed network significantly outperforms state-of-the-art methods on certain metrics and confirms the robustness and accuracy of our approach.

Details DOI

AAAI Conference 2025 Conference Paper

Spiking Point Transformer for Point Cloud Classification

Peixi Wu
Bosong Chai
Hebei Li
Menghua Zheng
Yansong Peng
Zeyu Wang
Xuan Nie
Yueyi Zhang

Spiking Neural Networks (SNNs) offer an attractive and energy-efficient alternative to conventional Artificial Neural Networks (ANNs) due to their sparse binary activation. When SNN meets Transformer, it shows great potential in 2D image processing. However, their application for 3D point cloud remains underexplored. To this end, we present Spiking Point Transformer (SPT), the first transformer-based SNN framework for point cloud classification. Specifically, we first design Queue-Driven Sampling Direct Encoding for point cloud to reduce computational costs while retaining the most effective support points at each time step. We introduce the Hybrid Dynamics Integrate-and-Fire Neuron (HD-IF), designed to simulate selective neuron activation and reduce over-reliance on specific artificial neurons. SPT attains state-of-the-art results on three benchmark datasets that span both real-world and synthetic datasets in the SNN domain. Meanwhile, the theoretical energy consumption of SPT is at least 6.4x less than its ANN counterpart.

PDF Details DOI