Xiaohong Xiang

Chongqing University of Posts and Telecommunications

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Xin Deng

Chongqing University of Posts and Telecommunications

Fuyuan Zhang

Fuzhou University

Xiaoyu Ding

University of Technology Sydney

Jun Zhao

Nanyang Technological University

Qi Wang

Beijing University of Technology

Xianping Zeng

Chinese Academy of Sciences

Zufu Pan

Beibu Gulf University

Quan Li

Xinjiang University

Xiaohu Huang

Beibu Gulf University

Mohan Kankanhalli

National University of Singapore

Cooperative Institutions

Chongqing University of Posts and Telecommunications

Beibu Gulf University

Chinese Academy of Sciences

Harbin Engineering University

Heilongjiang University of Science and Technology

Guangdong University of Technology

State Grid Corporation of China (China)

Shandong Institute of Automation

Guizhou University

Zhejiang University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

MSG-CAM:Multi-scale inputs make a better visual interpretation of CNN networks

2022 IEEE International Conference on Multimedia and Expo (ICME) (2023)

Xiaohong Xiang Fuyuan Zhang Xin Deng Ke Hu

The visualization of deep learning models has been widely studied as an effective means of exploring the decision-making processes within these models. However, current visualization methods suffer from several limitations, such as low resolution and poor visualization of multiple occurrences of the same class. In this paper, we propose a novel visualization technique called MSG-CAM, which is an improvement on the existing Group-CAM method. Our method uses the feature maps and gradients of the last layer of the convolutional neural network to create masks through multi-scale enlargement of the original input image and fusion of the resulting feature maps and gradients. Through both qualitative and quantitative analysis, we have demonstrated that the saliency maps generated by our method are more reasonable and accurately reflect the internal decision-making processes of the neural network.

Feature (linguistics)

10.1109/icme55011.2023.00061

Cite

Citations (1)

Defect Identification for Mild Steel in Arc Welding Using Multi-Sensor and Neighborhood Rough Set Approach

Applied Sciences (2024)

Xianping Zeng Zhiqiang Feng Xiaohong Xiang Xin Li Xiaohu Huang

Welding technology plays a vital role in the manufacturing process of ships, automobiles, and aerospace vehicles because it directly impacts their operational safety and reliability. Hence, the development of an accurate system for identifying welding defects in arc welding is crucial to enhancing the quality of welding production. In this study, a defect recognition method combining the Neighborhood Rough Set (NRS) with the Dingo Optimization Algorithm Support Vector Machine (DOA-SVM) in a multisensory framework is proposed. The 195-dimensional decision-making system mentioned above was constructed to integrate multi-source information from molten pool images, welding current, and vibration signals. To optimize the system, it was further refined to a 12-dimensional decision-making setup through outlier processing and feature selection based on the Neighborhood Rough Set. Subsequently, the DOA-SVM is employed for detecting welding defects. Experimental results demonstrate a 98.98% accuracy rate in identifying welding defects using our model. Importantly, this method outperforms comparative techniques in terms of quickly and accurately identifying five common welding defects, thereby affirming its suitability for arc welding. The proposed method not only achieves high accuracy but also simplifies the model structure, enhances detection efficiency, and streamlines network training.

Robot welding

10.3390/app14124978

Cite

Citations (0)

WPG-CAM: A Novel Weighted Feature Fusion CAM Method Based on Information Entropy Using Pooling and Gaussian Upsampling

Communications in computer and information science (2024)

Feifei Zhang Xiaohong Xiang Fuyuan Zhang Jun Zhao

Pooling

Upsampling

10.1007/978-981-97-7001-4_1

Cite

Citations (0)

CEO: Identifying Overlapping Communities via Construction, Expansion and Optimization

Information Sciences (2022)

Xiaoyu Ding Hailu Yang Jianpei Zhang Jing Yang Xiaohong Xiang

Optimization algorithm

10.1016/j.ins.2022.03.012

Cite

Citations (10)

Affect-based adaptive presentation of home videos

Proceedings of the 30th ACM International Conference on Multimedia (2011)

Xiaohong Xiang Mohan Kankanhalli

In recent times, the proliferation of multimedia devices and reduced costs of data storage have enabled people to easily record and collect a large number of home videos; furthermore, this collection is growing with time. With the popularity of participatory media such as YouTube and facebook, problems are encountered when people intend to share their home videos with others. The first problem is that different people might be interested in different video content. Given the numbers of home videos, it is a time-consuming and hard task to manually select proper content for people with different interests. Secondly, as short videos are becoming more and more popular in media sharing applications, people need to manually cut and edit home videos which is again a tedious task. In this paper, we propose a method that employs affective analysis to automatically create video presentations from home videos. Our novel method adaptively creates presentations based on three properties: emotional tone, local main character and global main character. A novel sparsity-based affective labeling method is proposed to identify the emotional content of the videos. The local and global main characters are determined by applying face recognition in each shot. To demonstrate the proposed method, three kinds of presentations are created for family, acquaintance and outsider. Experimental results show that our method is very effective in video sharing and the users are satisfied with the videos generated by our method.

Popularity

Presentation (obstetrics)

Tone (literature)

Affect

10.1145/2072298.2072370

Cite

Citations (13)

ML-TFN: Multi Layers Tensor Fusion Network for Affective Video Content Analysis

Communications in computer and information science (2022)

Qi Wang Xiaohong Xiang Jun Zhao

Modality (human–computer interaction)

Modalities

10.1007/978-981-19-6142-7_14

Cite

Citations (0)

Visualization Enhancement of Saliency Methods Based on the Sliding Window Mechanism

Frontiers in artificial intelligence and applications (2023)

Xiaohong Xiang Fuyuan Zhang Xin Deng Xiaoyu Ding

Deep neural networks are widely used in image classification tasks, but their internal decision-making mechanisms are often difficult to explain. While various algorithms have been developed to visualize these mechanisms, many of them produce coarse, noisy results that are not always convincing. To address this issue, we propose a method for enhancing saliency maps produced by saliency methods. Our method uses a fixed-size sliding window to upsample local regions of the input image and feed them into the selected visualization algorithm to generate class-specific saliency maps and probability scores. We then downsample the resulting saliency maps and multiply them by the probability scores to obtain maps with greater detail. We evaluate our method using different saliency methods and network architectures, and demonstrate its effectiveness through both quantitative metrics and intuitive evaluation. Our results show that our method significantly improves the performance of these saliency methods, providing a more valid and reliable means of visualizing the decision mechanisms of deep neural networks. Code is available at https://github.com/LuoLogic/Enhuncement-saliency.

Sliding window protocol

Code (set theory)

10.3233/faia230582

Cite

Citations (0)

DBCERT: Reconstruct the BERT Model for Chinese Spelling Correction

Rongpeng Li Xiaohong Xiang Bo Hu Xin Deng Jun Zhao

Spelling

10.1109/ispce-asia64773.2024.10756270

Cite

Citations (0)

Item-Behavior Sequence Session-Based Recommendation