XWXu Wang

Xu Wang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xu Wang has a total of 149 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • MDL-DA Track: Self-supervised Sperm Motility Direction Learning and Density-Adaptive Association for Sperm Tracking

    by Xiaoyu Zeng, Yang Zhao, Xu Wang, Longlong Fu, Jiahui Wu, Jihong Pei, Xuan Yang

  • Cell-Division-Aware Category-Enhanced Contrastive Learning Model for Embryonic Cleavage Stage Classification

    by Yukun Li, Yang Zhao, Xu Wang, Longlong Fu, Jiahui Wu, Xuan Yang, Jihong Pei

  • Interactive $360 ^{\circ }$ Object Removal: A Coordinate Querying Approach via Spherical Normal Constraints and Learnable Wavelet Enhancement

    by Qiudan Zhang, Ni Chen, Xuexun Liu, Jiashu Liu, Xu Wang, Jianmin Jiang

    Traditional perspective image inpainting tasks focus on restoring regions using a random mask image. However, the emergence of virtual reality (VR) renders traditional image inpainting inadequate for interactive editing, as users demand the ability to precisely remove specific objects in VR scenes rather than relying on arbitrary masks. Moreover, existing perspective inpainting methods exhibit inherent weaknesses in maintaining geometric distortions and photometric consistency in spherical projections. In this paper, we introduce a new task for interactively removing a queried object and subsequently repairing the missing regions in indoor $360^{\circ }$ images. Moreover, we propose a robust baseline methodology for indoor $360^{\circ }$ object removal that integrates coordinate querying with spherical normal constraints and learnable wavelet transform to enhance the three-dimensionality and lighting fidelity of repaired $360^{\circ }$ images. Specifically, we develop a spherical normal estimation module that takes advantage of the spherical projection properties of $360^{\circ }$ images to generate accurate surface normals for constraining the spherical geometry of the restored $360^{\circ }$ images. Subsequently, we design a wavelet-enhanced spatial feature aggregation module to improve directional continuity and frequency characteristics in restored regions, enabling the repaired $360 ^{\circ }$ images to exhibit clearer textures and sharper edges. Furthermore, to address the high-resolution nature of $360^{\circ }$ images, a mask region pixel coordinate querying module is introduced to alleviate memory constraints inherent in explicit $360 ^{\circ }$ image representations, enabling high-resolution content synthesis for any designated region. We also establish a new object removal dataset for indoor $360^{\circ }$ images (S3D-Removal), comprising 18k+ pairs of $360^{\circ }$ images with corresponding query points. Extensive experimental results on the S3D-Removal dataset demonstrate that our proposed method consistently outperforms state-of-the-art inpainting techniques across all evaluation metrics.

  • Rate-Reconfigurable Deep Point Cloud Compression With Perceptual Bit Allocation Optimization

    by Yun Zhang, Lewen Fan, Zixi Guo, Xu Wang, Xiaoxia Huang, Sam Kwong

    Conventional end-to-end learning-based point cloud compression requires training multiple models to adapt to different target bit rates. Moreover, the rate difference between geometry and attribute components of point clouds is not well-considered. In this paper, we propose an end-to-end Rate-Reconfigurable Deep Point Cloud Compression (RR-DPCC) with on/off-line Perceptual Bit Allocation Optimization (PBAO-ON/ OFF) , which achieves arbitrary bit rate control with one trained deep model and high efficiency joint geometry and attribute coding. First, we propose the framework of the RR-DPCC using PBAO-ON/OFF, which includes Point Cloud Quality Assessment (PCQA) for perceptual quality measurement, PBAO-ON/OFF modules for bit allocation and RR-DPCC for high efficiency point cloud coding. Second, we propose a one-stream network of the RR-DPCC to encode the attribute and geometry of point clouds jointly. Moreover, in RR-DPCC, a bitrate reconfigurable module is proposed to encode multiple fine-grained bitrate points with one trained model and a rate allocation module is proposed to allocate bits between geometry and attribute. Third, we propose on/off-line PBAO algorithms to maximize the perceptual quality of the reconstructed point cloud, where the bits are properly allocated based on the importance of geometry and attribute. Meanwhile, rate-distortion models (R-α/β and D-α/β) are derived for high accuracy rate control and bit allocation. Experimental results show that the proposed RR-DPCC achieves fine-grained bitrate control and allocation through a single trained model. When combined the proposed RR-DPCC with PBAO-ON, it reduces -6.56% and -18.68% bit rate on average as comparing with the state-of-the-art V-PCC and Deep Joint Geometry and Attribute Compression (Deep-JGAC), respectively. When combined with the PBAO-OFF, it reduces - 4.90% and -15.34% bit rate reductions on average, and reduces 98.38%/22.05% and 53.75%/10.04% encoding/decoding time on average with respect to V-PCC and Deep-JGAC.


  • A Visual-Linguistic Approach for Robust RGB-Thermal Tracking with Dynamic Template Adaptation

    by Xu Wang, Huanxin Zheng, Haohong Liao, Qiudan Zhang, Lin Ma, Jianmin Jiang

    RGB-T tracking seeks to improve tracking robust ness in complex environments by exploiting the complementary advantages of RGB and thermal infrared (TIR) modalities. Nevertheless, current RGB-T tracking methods encounter two critical limitations stemming from their reliance on fixed template images for search region matching. First, the resemblance between the initial fixed template and the target in later frames diminishes as time goes on, causing tracking reliability to decline. Second, background clutter within template bounding boxes introduces disruptive noise that compromises tracking accuracy. To address these challenges, we propose a new referring RGB-T tracking approach that integrates visual-linguistic cues to enhance multimodal data, enabling background noise elimination in the template and dynamic template image updating through adaptive mechanisms. Additionally, to facilitate research in language guided multimodal tracking, we construct a large-scale dataset named Refer-RGBT, which comprises 1508 multimodal video pairs with synchronized RGB/TIR sequences of diverse objects. Each sequence is annotated by descriptive textual captions that detail their appearance and actions. Extensive evaluations on the Refer-RGBT dataset demonstrate that our proposed approach achieves cutting-edge performance compared to the state-of-the art (SOTA) methods.


  • Top Cited Papers

    Deep Learning-Based Joint Geometry and Attribute Up-Sampling for Large-Scale Colored Point Clouds

    by Yun Zhang, Feifan Chen, Na Li, Zhiwei Guo, Xu Wang, Fen Miao, Sam Kwong

    Colored point cloud comprising geometry and attribute components is one of the mainstream representations enabling realistic and immersive 3D applications. To generate large-scale and denser colored point clouds, we propose a deep learning-based Joint Geometry and Attribute Up-sampling (JGAU) method, which learns to model both geometry and attribute patterns and leverages the spatial attribute correlation. Firstly, we establish and release a large-scale dataset for colored point cloud up-sampling, named SYSU-PCUD, which has 121 large-scale colored point clouds with diverse geometry and attribute complexities in six categories and four sampling rates. Secondly, to improve the quality of up-sampled point clouds, we propose a deep learning-based JGAU framework to up-sample the geometry and attribute jointly. It consists of a geometry upsampling network and an attribute up-sampling network, where the latter leverages the up-sampled auxiliary geometry to model neighborhood correlations of the attributes. Thirdly, we propose two coarse attribute up-sampling methods, Geometric Distance Weighted Attribute Interpolation (GDWAI) and Deep Learning-based Attribute Interpolation (DLAI), to generate coarsely upsampled attributes for each point. Then, we propose an attribute enhancement module to refine the up-sampled attributes and generate high quality point clouds by further exploiting intrinsic attribute and geometry patterns. Extensive experiments show that Peak Signal-to-Noise Ratio (PSNR) achieved by the proposed JGAU are 33.90 dB, 32.10 dB, 31.10 dB, and 30.39 dB when upsampling rates are 4×, 8×, 12×, and 16×, respectively. Compared to the state-of-the-art schemes, the JGAU achieves an average of 2.32 dB, 2.47 dB, 2.28 dB and 2.11 dB PSNR gains at four upsampling rates, respectively, which are significant.


  • Open Access

    3D Weakly Supervised Semantic Segmentation via Class-Aware and Geometry-Guided Pseudo-Label Refinement

    by Xiaoxu Xu, Xuexun Liu, Jinlong Li, Yitian Yuan, Qiudan Zhang, Lin Ma, Nicu Sebe, Xu Wang

    3D weakly supervised semantic segmentation (3D WSSS) aims to achieve semantic segmentation by leveraging sparse or low-cost annotated data, significantly reducing reliance on dense point-wise annotations. Previous works mainly employ class activation maps or pre-trained vision-language models to address this challenge. However, the low quality of pseudo-labels and the insufficient exploitation of 3D geometric priors jointly create significant technical bottlenecks in developing high-performance 3D WSSS models. In this paper, we propose a simple yet effective 3D weakly supervised semantic segmentation method that integrates 3D geometric priors into a class-aware guidance mechanism to generate high-fidelity pseudo labels. Concretely, our designed methodology first employs Class-Aware Label Refinement module to generate more balanced and accurate pseudo labels for semantic categrories. This initial refinement stage focuses on enhancing label quality through category-specific optimization. Subsequently, the Geometry-Aware Label Refinement component is developed, which strategically integrates implicit 3D geometric constraints to effectively filter out low-confidence pseudo labels that fail to comply with geometric plausibility. Moreover, to address the challenge of extensive unlabeled regions, we propose a Label Update strategy that integrates Self-Training to propagate labels into these areas. This iterative process continuously enhances pseudo-label quality while expanding label coverage, ultimately fostering the development of high-performance 3D WSSS models. Comprehensive experimental validation reveals that our proposed methodology achieves state-of-the-art performance on both ScanNet and S3DIS benchmarks while demonstrating remarkable generalization capability in unsupervised settings, maintaining competitive accuracy through its robust design.