Jian Chu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jian Chu has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Genke Yang
Shanghai Jiao Tong University
14 shared publications - Chao Xu
Zhejiang University
7 shared publications - Hongye Su
Zhejiang University
4 shared publications - Shengze Cai
College of Control Science and Engineering, Zhejiang University, Hangzhou, China
4 shared publications - Xinxing Yang
Department of Biomedical Engineering, Oregon Health & Science University, Portland, OR, USA
4 shared publications - Fei Gao
Zhejiang University
3 shared publications
Latest Publications
- 2 August 2026
- Published by Elsevier in Neurocomputing
- Vol. 702
PHA-Net: Prototype-based hierarchical alignment network for text-video retrieval
- 2 August 2026
- Published by Elsevier in Neurocomputing
- Vol. 702
PHA-Net: Prototype-based hierarchical alignment network for text-video retrieval
- 29 January 2026
- Published by Elsevier in International Journal of Hydrogen Energy
- Vol. 212
Process integration and optimization of green ammonia synthesis under fluctuating renewable energy supply
- 29 January 2026
- Published by Elsevier in International Journal of Hydrogen Energy
- Vol. 212
Process integration and optimization of green ammonia synthesis under fluctuating renewable energy supply
- 10 December 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 170
HPGoG: Hetero-pooling integrated with graph-of-graphs to alleviate class imbalance for fault tracing of transformer substation devices
- 10 December 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 170
HPGoG: Hetero-pooling integrated with graph-of-graphs to alleviate class imbalance for fault tracing of transformer substation devices
- 9 December 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Image and Graphics
Chinese Text Image Super-Resolution Network Using SwinIR and Text-Aware Loss
Scene Text Image Super-Resolution (STISR) is a challenging task in computer vision. Existing STISR methods primarily focus on English characters, with insufficient attention given to the more complex Chinese characters. To address this, we propose SwinTISR, an STISR method based on SwinIR, specifically designed for Chinese text images. To better reconstruct the intricate stroke structures of Chinese characters, we introduce a text-aware loss [ ] that combines stroke-focused features and text location guidance. Additionally, Canny edge maps are utilized as auxiliary supervision to further enhance the visual quality and readability. Extensive experiments on the STISR benchmarks TextZoom and Real-CE demonstrate the effectiveness and superiority of the proposed SwinTISR. Furthermore, we enhance the Chinese STISR dataset Real-CE by incorporating the Multi-scale Homography Network.
- 9 December 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Image and Graphics
Chinese Text Image Super-Resolution Network Using SwinIR and Text-Aware Loss
Scene Text Image Super-Resolution (STISR) is a challenging task in computer vision. Existing STISR methods primarily focus on English characters, with insufficient attention given to the more complex Chinese characters. To address this, we propose SwinTISR, an STISR method based on SwinIR, specifically designed for Chinese text images. To better reconstruct the intricate stroke structures of Chinese characters, we introduce a text-aware loss [ ] that combines stroke-focused features and text location guidance. Additionally, Canny edge maps are utilized as auxiliary supervision to further enhance the visual quality and readability. Extensive experiments on the STISR benchmarks TextZoom and Real-CE demonstrate the effectiveness and superiority of the proposed SwinTISR. Furthermore, we enhance the Chinese STISR dataset Real-CE by incorporating the Multi-scale Homography Network.
- 21 June 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Pattern Recognition and Artificial Intelligence
- Vol. 39 (10)
Image Super-Resolution with Dense-Parallel Swin Transformer
Recent advancements in Swin Transformer-based image super-resolution (SR) methods have demonstrated remarkable progress through efficient modeling capabilities and feature extraction. However, existing approaches still have limitations: in hierarchical feature representation architectures, inter-layer feature propagation efficiency progressively degrades with increasing network depth. For this, the paper proposes an SR model, namely, SwinDPSR, composed of Dense-Parallel Swin Transformer (DPST) blocks. The proposed method innovatively constructs a densely connected parallel Swin Transformer architecture that enhances information flow through multi-level feature reuse mechanisms. Simultaneously, we introduce the Spatial Frequency Block (SFB) that establishes global contextual correlations in the frequency domain branch to precisely compensate for local detail loss. Experiments demonstrate significant improvements over the SwinIR across five benchmarks (Set5, Set14, BSD100, Urban100, and Manga109), achieving an average PSNR gain of 0.28[ ]dB in [ ] SR tasks. Theoretical analysis and experimental validation confirm the effectiveness of the parallel architecture, which improves FPS by 6.8% through enhanced computational parallelism. Ablation studies verify the synergistic optimization effect of dense connections and SFB.
- 21 June 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Pattern Recognition and Artificial Intelligence
- Vol. 39 (10)
Image Super-Resolution with Dense-Parallel Swin Transformer
Recent advancements in Swin Transformer-based image super-resolution (SR) methods have demonstrated remarkable progress through efficient modeling capabilities and feature extraction. However, existing approaches still have limitations: in hierarchical feature representation architectures, inter-layer feature propagation efficiency progressively degrades with increasing network depth. For this, the paper proposes an SR model, namely, SwinDPSR, composed of Dense-Parallel Swin Transformer (DPST) blocks. The proposed method innovatively constructs a densely connected parallel Swin Transformer architecture that enhances information flow through multi-level feature reuse mechanisms. Simultaneously, we introduce the Spatial Frequency Block (SFB) that establishes global contextual correlations in the frequency domain branch to precisely compensate for local detail loss. Experiments demonstrate significant improvements over the SwinIR across five benchmarks (Set5, Set14, BSD100, Urban100, and Manga109), achieving an average PSNR gain of 0.28[ ]dB in [ ] SR tasks. Theoretical analysis and experimental validation confirm the effectiveness of the parallel architecture, which improves FPS by 6.8% through enhanced computational parallelism. Ablation studies verify the synergistic optimization effect of dense connections and SFB.
- 9 May 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Pattern Recognition and Artificial Intelligence
- Vol. 39 (06)
Bayesian Uncertainty Weighted Optimization for Offline Reinforcement Learning
Offline reinforcement learning (offline RL) endeavors to learn effective policies from a large batch of pre-collected datasets without any costly or dangerous online exploration. Nevertheless, offline RL always suffers from substantial algorithmic extrapolation errors and may fail when bootstrapping from out-of-distribution (OOD) actions or states. In this work, we introduce a practical and effective Bayesian uncertainty weighted optimization (BUWO) to leverage the Bayesian uncertainty to account for the epistemic uncertainty associated with each training sample and penalize the state-action pairs with high uncertainty. We compare BUWO with other prevailing offline RL algorithms on D4RL benchmarks. The experimental results demonstrate that the algorithm can enhance the average reward score by almost 15% without additional computational costs compared to the current state-of-the-art algorithm.
- 9 May 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Pattern Recognition and Artificial Intelligence
- Vol. 39 (06)
Bayesian Uncertainty Weighted Optimization for Offline Reinforcement Learning
Offline reinforcement learning (offline RL) endeavors to learn effective policies from a large batch of pre-collected datasets without any costly or dangerous online exploration. Nevertheless, offline RL always suffers from substantial algorithmic extrapolation errors and may fail when bootstrapping from out-of-distribution (OOD) actions or states. In this work, we introduce a practical and effective Bayesian uncertainty weighted optimization (BUWO) to leverage the Bayesian uncertainty to account for the epistemic uncertainty associated with each training sample and penalize the state-action pairs with high uncertainty. We compare BUWO with other prevailing offline RL algorithms on D4RL benchmarks. The experimental results demonstrate that the algorithm can enhance the average reward score by almost 15% without additional computational costs compared to the current state-of-the-art algorithm.
- Open Access
TC-MGC: Text-Conditioned Multi-Grained Contrastive Learning for Text-Video Retrieval
Motivated by the success of coarse-grained or fine-grained contrast in text-video retrieval, there emerge multi-grained contrastive learning methods which focus on the integration of contrasts with different granularity. However, due to the wider semantic range of videos, the text-agnostic video representations might encode misleading information not described in texts, thus impeding the model from capturing precise cross-modal semantic correspondence. To this end, we propose a Text-Conditioned Multi-Grained Contrast framework, dubbed TC-MGC. Specifically, our model employs a language-video attention block to generate aggregated frame and video representations conditioned on the word's and text's attention weights over frames. To filter unnecessary similarity interactions and decrease trainable parameters in the Interactive Similarity Aggregation (ISA) module, we design a Similarity Reorganization (SR) module to identify attentive similarities and reorganize cross-modal similarity vectors and matrices. Next, we argue that the imbalance problem among multigrained similarities may result in over- and under-representation issues. We thereby introduce an auxiliary Similarity Decorrelation Regularization (SDR) loss to facilitate cooperative relationship utilization by similarity variance minimization on matching text-video pairs. Finally, we present a Linear Softmax Aggregation (LSA) module to explicitly encourage the interactions between multiple similarities and promote the usage of multi-grained information. Empirically, TC-MGC achieves competitive results on multiple text-video retrieval benchmarks, outperforming X-CLIP model by +2.8% (+1.3%), +2.2% (+1.0%), +1.5% (+0.9%) relative (absolute) improvements in text-to-video retrieval R@1 on MSR-VTT, DiDeMo and VATEX, respectively. Our code is publicly available at https://github.com/JingXiaolun/TC-MGC.
Open AccessTC-MGC: Text-Conditioned Multi-Grained Contrastive Learning for Text-Video Retrieval
Motivated by the success of coarse-grained or fine-grained contrast in text-video retrieval, there emerge multi-grained contrastive learning methods which focus on the integration of contrasts with different granularity. However, due to the wider semantic range of videos, the text-agnostic video representations might encode misleading information not described in texts, thus impeding the model from capturing precise cross-modal semantic correspondence. To this end, we propose a Text-Conditioned Multi-Grained Contrast framework, dubbed TC-MGC. Specifically, our model employs a language-video attention block to generate aggregated frame and video representations conditioned on the word's and text's attention weights over frames. To filter unnecessary similarity interactions and decrease trainable parameters in the Interactive Similarity Aggregation (ISA) module, we design a Similarity Reorganization (SR) module to identify attentive similarities and reorganize cross-modal similarity vectors and matrices. Next, we argue that the imbalance problem among multigrained similarities may result in over- and under-representation issues. We thereby introduce an auxiliary Similarity Decorrelation Regularization (SDR) loss to facilitate cooperative relationship utilization by similarity variance minimization on matching text-video pairs. Finally, we present a Linear Softmax Aggregation (LSA) module to explicitly encourage the interactions between multiple similarities and promote the usage of multi-grained information. Empirically, TC-MGC achieves competitive results on multiple text-video retrieval benchmarks, outperforming X-CLIP model by +2.8% (+1.3%), +2.2% (+1.0%), +1.5% (+0.9%) relative (absolute) improvements in text-to-video retrieval R@1 on MSR-VTT, DiDeMo and VATEX, respectively. Our code is publicly available at https://github.com/JingXiaolun/TC-MGC.
- 26 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 70 (9) , 5792-5807
Accelerated Distributed Aggregative Optimization
This article delves into the investigation of a distributed aggregative optimization problem within a network. In this scenario, each agent possesses its own local cost function, which relies not only on the local state variable but also on an aggregated function of state variables from all agents. To expedite the optimization process, we amalgamate the heavy ball and Nesterov's accelerated method with distributed aggregative gradient tracking, resulting in the proposal of two innovative algorithms, aimed at resolving the distributed aggregative optimization problem. Our analysis demonstrates that the proposed algorithms can converge to an optimal solution at a global linear convergence rate when the objective function is strongly convex with the Lipschitz-continuous gradient, and when the parameters (e.g., step size and momentum coefficients) are chosen within specific ranges. In addition, we present several numerical experiments to verify the effectiveness, robustness, and superiority of our proposed algorithms.
- 26 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 70 (9) , 5792-5807
Accelerated Distributed Aggregative Optimization
This article delves into the investigation of a distributed aggregative optimization problem within a network. In this scenario, each agent possesses its own local cost function, which relies not only on the local state variable but also on an aggregated function of state variables from all agents. To expedite the optimization process, we amalgamate the heavy ball and Nesterov's accelerated method with distributed aggregative gradient tracking, resulting in the proposal of two innovative algorithms, aimed at resolving the distributed aggregative optimization problem. Our analysis demonstrates that the proposed algorithms can converge to an optimal solution at a global linear convergence rate when the objective function is strongly convex with the Lipschitz-continuous gradient, and when the parameters (e.g., step size and momentum coefficients) are chosen within specific ranges. In addition, we present several numerical experiments to verify the effectiveness, robustness, and superiority of our proposed algorithms.