Weizhi Xian
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Weizhi Xian has a total of 112 co-authors. Here are the top co-authors by number of shared publications:
- Xuekai Wei
School of Computer Science, Chongqing University, Chongqing 400044, China
24 shared publications - Mingliang Zhou
School of Computer Science, Chongqing University, Chongqing, China
21 shared publications - Jielu Yan
School of Computer Science, Chongqing University, Chongqing 400044, China
20 shared publications - Bin Fang
College of Computer Science, Chongqing University, Chongqing 400044, P. R. China
8 shared publications - Bin Chen
Chongqing Research Institute of Harbin Institute of Technology, Harbin Institute of Technology, Chongqing 401151, China
6 shared publications - Jun Luo
School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences, Beijing, China
6 shared publications
Latest Publications
- Open Access
- 10 November 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (22) , 11119
Multi-Horizon AQI Forecasting Through the Heterogeneous Fusion of a Temporal Convolutional Network and a Graph-Conditioned Causal Transformer
Recent station observations provide a useful basis for air quality index (AQI) forecasting, whereas regional measurements may supply additional context at longer forecast leads. In this study, the forecasts of an independently trained temporal convolutional network (TCN) and a graph-conditioned causal transformer (Graph-CT) are combined. Both use the same input window of 24 h to predict all the hourly leads from 1 to 24 h, and their corresponding outputs are averaged. Air-quality records from 26 stations in the Chengdu–Chongqing economic circle are reconstructed on a common hourly grid, with causal input filling and evaluation restricted to observed targets. Learning rates and training durations are selected on 2021 data, followed by retrospective evaluation on the previously inspected 2022 benchmark. Across five runs, equal fusion achieves a pooled root mean square error (RMSE) of 18.036 ± 0.103 AQI, compared with 19.491 ± 0.063 for TCN, 18.284 ± 0.140 for Graph-CT, 19.413 ± 0.174 for iTransformer, and 19.043 ± 0.455 for Graph WaveNet under the common task. The paired interval includes zero, leaving the advantage over a two-member Graph-CT ensemble unresolved. Validation calibration slightly decreases the RMSE but increases the mean absolute error (MAE), and removing the transformer branch improves the evaluated configuration. Forecast averaging is therefore useful on this benchmark, although the results favour neither equal coefficients as an optimum nor the full architecture over its simpler variants. The evaluation of untouched periods and regions remains necessary for assessing generalization.
- 10 November 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (22) , 11119
Open AccessMulti-Horizon AQI Forecasting Through the Heterogeneous Fusion of a Temporal Convolutional Network and a Graph-Conditioned Causal Transformer
Recent station observations provide a useful basis for air quality index (AQI) forecasting, whereas regional measurements may supply additional context at longer forecast leads. In this study, the forecasts of an independently trained temporal convolutional network (TCN) and a graph-conditioned causal transformer (Graph-CT) are combined. Both use the same input window of 24 h to predict all the hourly leads from 1 to 24 h, and their corresponding outputs are averaged. Air-quality records from 26 stations in the Chengdu–Chongqing economic circle are reconstructed on a common hourly grid, with causal input filling and evaluation restricted to observed targets. Learning rates and training durations are selected on 2021 data, followed by retrospective evaluation on the previously inspected 2022 benchmark. Across five runs, equal fusion achieves a pooled root mean square error (RMSE) of 18.036 ± 0.103 AQI, compared with 19.491 ± 0.063 for TCN, 18.284 ± 0.140 for Graph-CT, 19.413 ± 0.174 for iTransformer, and 19.043 ± 0.455 for Graph WaveNet under the common task. The paired interval includes zero, leaving the advantage over a two-member Graph-CT ensemble unresolved. Validation calibration slightly decreases the RMSE but increases the mean absolute error (MAE), and removing the transformer branch improves the evaluated configuration. Forecast averaging is therefore useful on this benchmark, although the results favour neither equal coefficients as an optimum nor the full architecture over its simpler variants. The evaluation of untouched periods and regions remains necessary for assessing generalization.
- Open Access
- 8 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (10) , 1-20
FR-IQA: Joint Perceptual Degradation in Collaborative Feature Refinement
A full-reference image quality assessment (FR-IQA) is proposed in this work. First, images are transformed into wavelet subbands and color histograms to maintain key visual characteristics of the human visual system (HVS). Second, subtle degradations are amplified via collaborative refinement. Afterward, the Hausdorff distance is used to capture local geometric discrepancies, rendering the measurement robust to severe artifacts and distortions. The experimental results demonstrate that the proposed approach accurately predicts subjective quality scores. Aligning more closely with HVS, the proposed method outperforms existing state-of-the-art methods on multiple benchmark databases. The code is available at https://github.com/OC007/IQACODE2025 .
- 8 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (10) , 1-20
Open AccessFR-IQA: Joint Perceptual Degradation in Collaborative Feature Refinement
A full-reference image quality assessment (FR-IQA) is proposed in this work. First, images are transformed into wavelet subbands and color histograms to maintain key visual characteristics of the human visual system (HVS). Second, subtle degradations are amplified via collaborative refinement. Afterward, the Hausdorff distance is used to capture local geometric discrepancies, rendering the measurement robust to severe artifacts and distortions. The experimental results demonstrate that the proposed approach accurately predicts subjective quality scores. Aligning more closely with HVS, the proposed method outperforms existing state-of-the-art methods on multiple benchmark databases. The code is available at https://github.com/OC007/IQACODE2025 .
- 29 July 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 333
Semantic-guided multi-feature fusion for underwater image quality assessment
- 29 July 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 333
Semantic-guided multi-feature fusion for underwater image quality assessment
- Open Access
- 30 March 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (7) , 3345
IM-DETR: DETR with Mix-Encoder for Industrial Scenarios
Industrial defect detection is a fundamental task in intelligent manufacturing, yet existing object detection methods often struggle with the characteristics of industrial defects, such as small size, irregular shapes, and complex visual backgrounds. Moreover, most detection models are designed primarily for natural image datasets, resulting in limited robustness when deployed in real-world industrial environments. To address these challenges, this research focuses on industrial defect detection and presents contributions at both the dataset and method levels. First, two real-world industrial defect datasets collected from actual production lines are introduced, namely, the Stator Housing Defect Dataset and the Cover Plate Silicone Defect Dataset, which cover representative inspection scenarios with distinct defect characteristics. Second, we propose a detection transformer with a mixed encoder for industrial scenarios (IM-DETR). By integrating heterogeneous multi-scale feature representations, the proposed framework jointly enhances local detail sensitivity and global contextual reasoning without relying on complex post-processing. Extensive experiments on the proposed industrial datasets demonstrate that IM-DETR consistently outperforms existing state-of-the-art detection methods, particularly in scenarios involving small defects, complex backgrounds, and appearance ambiguity, validating the effectiveness and robustness of the proposed approach.
- 30 March 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (7) , 3345
Open AccessIM-DETR: DETR with Mix-Encoder for Industrial Scenarios
Industrial defect detection is a fundamental task in intelligent manufacturing, yet existing object detection methods often struggle with the characteristics of industrial defects, such as small size, irregular shapes, and complex visual backgrounds. Moreover, most detection models are designed primarily for natural image datasets, resulting in limited robustness when deployed in real-world industrial environments. To address these challenges, this research focuses on industrial defect detection and presents contributions at both the dataset and method levels. First, two real-world industrial defect datasets collected from actual production lines are introduced, namely, the Stator Housing Defect Dataset and the Cover Plate Silicone Defect Dataset, which cover representative inspection scenarios with distinct defect characteristics. Second, we propose a detection transformer with a mixed encoder for industrial scenarios (IM-DETR). By integrating heterogeneous multi-scale feature representations, the proposed framework jointly enhances local detail sensitivity and global contextual reasoning without relying on complex post-processing. Extensive experiments on the proposed industrial datasets demonstrate that IM-DETR consistently outperforms existing state-of-the-art detection methods, particularly in scenarios involving small defects, complex backgrounds, and appearance ambiguity, validating the effectiveness and robustness of the proposed approach.
- Open Access
- 27 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (3) , 1-20
Neighborhood Attention-based Feature Reconstruction for Image Anomaly Detection and Localization
With the advancement of machine vision technology, automated vision inspection systems are needed in broad quality control scenarios. This article proposes a neighborhood attention-based feature reconstruction method for image anomaly detection and localization (NAFRAD). To address the challenges of data scarcity, low visibility, and irregular defect shapes in unsupervised anomaly detection, we introduce a feature reconstruction framework that preserves high-level abstract features rather than focusing on pixel-level reconstruction. This approach enhances model robustness and generalizability by leveraging neighborhood attention (NA) mechanisms, which simultaneously capture local details and the global context through a sliding window strategy. The NA-based autoencoder reconstructs normal features by aggregating local inductive biases with translational equivariance, enabling precise anomaly localization. Extensive experiments on the MVTec Anomaly Detection (MVTec AD) dataset—comprising 15 categories with 5,354 images—demonstrate the superiority of NAFRAD. It achieves state-of-the-art performance with AUROC \({}_{I}\) = 99.02, AUROC \({}_{P}\) = 98.99, and AP = 79.40, outperforming existing methods by 3.6% in AP and 0.89% in AUROC \({}_{P}\) . The framework’s effectiveness is validated through ablation studies, visualization of feature reconstruction, and comparisons with eight leading unsupervised methods. The code is made public at https://github.com/Math-Computer/NAFRAD .- 27 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (3) , 1-20
Open AccessNeighborhood Attention-based Feature Reconstruction for Image Anomaly Detection and Localization
With the advancement of machine vision technology, automated vision inspection systems are needed in broad quality control scenarios. This article proposes a neighborhood attention-based feature reconstruction method for image anomaly detection and localization (NAFRAD). To address the challenges of data scarcity, low visibility, and irregular defect shapes in unsupervised anomaly detection, we introduce a feature reconstruction framework that preserves high-level abstract features rather than focusing on pixel-level reconstruction. This approach enhances model robustness and generalizability by leveraging neighborhood attention (NA) mechanisms, which simultaneously capture local details and the global context through a sliding window strategy. The NA-based autoencoder reconstructs normal features by aggregating local inductive biases with translational equivariance, enabling precise anomaly localization. Extensive experiments on the MVTec Anomaly Detection (MVTec AD) dataset—comprising 15 categories with 5,354 images—demonstrate the superiority of NAFRAD. It achieves state-of-the-art performance with AUROC \({}_{I}\) = 99.02, AUROC \({}_{P}\) = 98.99, and AP = 79.40, outperforming existing methods by 3.6% in AP and 0.89% in AUROC \({}_{P}\) . The framework’s effectiveness is validated through ablation studies, visualization of feature reconstruction, and comparisons with eight leading unsupervised methods. The code is made public at https://github.com/Math-Computer/NAFRAD . - 26 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Broadcasting
- Vol. 72 (2) , 570-582
Image Quality Assessment: Unifying Local and Global Distortions via Fisher Information Geometry and Optimal Sinkhorn Transport
A prevailing limitation of existing full-reference image quality assessment (FR-IQA) methods is their segregated focus on either local texture or global structure, preventing a unified analysis of these two critical and interrelated dimensions of human perception. To resolve this, we propose the Fisher Sinkhorn image quality assessment (FSIQA) framework, which establishes a principled unification of information geometry and optimal transport theory. FSIQA leverages a dual-branch architecture: its Fisher branch captures localized, nonGaussian texture distortions by modelling higher-order statistics on a Riemannian manifold, whereas the complementary Sinkhorn branch assesses global structural deformations via entropy-regularized optimal transport. This synergistic framework creates a mathematical continuum for local-to-global degradation analysis and is notably independent of the backbone network architecture, thus requiring no dataset-specific retraining. Comprehensive experiments across five benchmark datasets validate the efficacy of our approach, demonstrating that FSIQA consistently outperforms existing methods and establishes a new state-of-the-art method in the field.
- 26 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Broadcasting
- Vol. 72 (2) , 570-582
Image Quality Assessment: Unifying Local and Global Distortions via Fisher Information Geometry and Optimal Sinkhorn Transport
A prevailing limitation of existing full-reference image quality assessment (FR-IQA) methods is their segregated focus on either local texture or global structure, preventing a unified analysis of these two critical and interrelated dimensions of human perception. To resolve this, we propose the Fisher Sinkhorn image quality assessment (FSIQA) framework, which establishes a principled unification of information geometry and optimal transport theory. FSIQA leverages a dual-branch architecture: its Fisher branch captures localized, nonGaussian texture distortions by modelling higher-order statistics on a Riemannian manifold, whereas the complementary Sinkhorn branch assesses global structural deformations via entropy-regularized optimal transport. This synergistic framework creates a mathematical continuum for local-to-global degradation analysis and is notably independent of the backbone network architecture, thus requiring no dataset-specific retraining. Comprehensive experiments across five benchmark datasets validate the efficacy of our approach, demonstrating that FSIQA consistently outperforms existing methods and establishes a new state-of-the-art method in the field.
- 13 February 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 338
SAM-IAD: Injecting specific knowledge into SAM for industrial anomaly detection
- 13 February 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 338
SAM-IAD: Injecting specific knowledge into SAM for industrial anomaly detection
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 13621-13634
PG-SWIQA: A Physics-Guided Spatial-Wavelet Interaction Network for No-Reference Underwater Image Quality Assessment
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 13621-13634
PG-SWIQA: A Physics-Guided Spatial-Wavelet Interaction Network for No-Reference Underwater Image Quality Assessment