14 March 2024 in IEEE Transactions on Multimedia
The object of image super-resolution reconstruction is to overcome the limitations imposed by hardware imaging conditions and patterns, aiming to restore high-frequency details in images through signal processing techniques. Recently, deep learning-based single-image super-resolution reconstruction (SISR) has achieved remarkable performance. However, the current methods exhibit inadequate performance in the reconstruction of texture details, thereby posing a challenge for further enhancing the accuracy of super-resolution reconstruction. In this study, we propose a novel contextual texture enhancement network (CTE-Net) aimed at improving the level of texture details in image super-resolution. The CTE-Net comprises of two crucial components: the multi-level feature aggregation module (MFAM) and the contextual information enhancement module (CIEM). The MFAM integrates global and local low-resolution (LR) features from both the pixel space and channel dimensions, thereby enhancing the feature representation capability of the network. The CIEM is deployed to enhance the network's learning capacity by integrating a meticulously designed context-attention mechanism, which effectively explores the adjacent contextual information of images and thereby amplifies the expressive capability of the generated features. Moreover, we utilize local binary patterns (LBP) to guide the feature selection strategies for MFAM and CIEM, thereby prioritizing the network's decision logic towards the recovery of texture details. The extensive experiments demonstrate that our method yields satisfactory results. In comparison to the state-of-the-art approaches, our method exhibits superior performance on the benchmark datasets.
17 January 2024 in Expert Systems with Applications
22 December 2025 in Aerospace Science and Technology
14 November 2025 in IEEE Transactions on Multimedia
Image inpainting has attracted considerable attention in computer vision and image processing due to its wide range of applications. While deep learning-based methods have shown promising potential, accurately recovering pixel-level details remains a significant challenge, particularly in the presence of large and irregular missing regions. Furthermore, existing methods are limited by unidirectional semantic guidance and a localized understanding of global structural context. In this study, we propose a mask-guided dual-branch Transformer-based framework, named MDT-FI, which effectively balances local detail restoration and global contextual reasoning by explicitly modeling long-range dependencies. MDT-FI consists of three key components: the Interactive Attention Module (IAM), the Spectral Harmonization Module (SHM), and the Lateral Adaptation Network (LAN). The model integrates multi-scale feature interaction, frequency-domain information fusion, and a mask-guided attention mechanism to progressively build cross-level feature associations. This design facilitates multi-level representation learning and optimization, thereby enhancing local texture synthesis while preserving global structural consistency. To further improve perceptual quality, a feature augmenter is employed to assess the fidelity of both texture and structure in the generated results. Extensive experiments on CelebA-HQ, Places2, and Paris Street View demonstrate that MDT-FI significantly outperforms state-of-the-art methods.
27 April 2026 in SSRN Electronic Journal
RGB-D salient object detection (SOD) relies on effective synergy between RGB and depth modalities for precise localization of salient objects. However, semantic inconsistencies between modalities weaken structural and edge information during fusion, destabilizing salient object representations across scales and limiting detection performance. To address this, we propose a cross-modal edge-guided consistency learning framework (CMCL-Net), stabilizing semantic distributions by jointly modeling cross-modal consistency and structural integrity. Specifically, the semantic coordination and consistency fusion (SCCF) module employs a content-aware weighting mechanism to adaptively align cross-modal features and enhance complementarity. The fine-grained edge guidance (FEGM) module leverages explicit edge supervision to strengthen structural perception of object boundaries, improving contour integrity and edge clarity. To handle large scale variations of salient objects, the multi-scale salient feature perception (MSFP) module integrates global semantic context with local details, enhancing multi-scale salient region modeling. Extensive experiments demonstrate that CMCL-Net outperforms state-of-the-art methods across multiple RGB-D SOD benchmarks.