中山大学智能媒体编码团队 logo 智能视觉编码团队Intelligent Visual Coding Team

Publications

学术成果

团队近三年的学术论文成果如下，更多论文可访问张云教授个人主页。

2026

论文代表图：Rate-Reconfigurable Deep Point Cloud Compression With Perceptual Bit Allocation Optimization
Rate-Reconfigurable Deep Point Cloud Compression With Perceptual Bit Allocation Optimization

Yun Zhang, Lewen Fan, Zixi Guo, Xu Wang, Xiaoxia Huang, Sam Kwong. IEEE Transactions on Image Processing, vol. 35, pp. 3451-3465, 2026.

简介：针对不同目标码率需要分别训练模型的问题，提出可重配置的 RR-DPCC，并设计在线/离线感知比特分配策略。该方法用单一模型实现灵活码率控制，同时协调点云几何与属性的编码开销。

papercode
论文代表图：Deep-JGAC: End-to-End Deep Joint Geometry and Attribute Compression for Dense Colored Point Clouds
Deep-JGAC: End-to-End Deep Joint Geometry and Attribute Compression for Dense Colored Point Clouds

Yun Zhang, Zhiwei Guo, Zixi Guo, Linwei Zhu, C.-C. Jay Kuo. IEEE Transactions on Circuits and Systems for Video Technology, 2026.

简介：面向稠密彩色点云，提出端到端几何与属性联合压缩框架 Deep-JGAC。方法通过残差自注意力几何编码、重着色与联合熵建模，提高两类信息的协同表示和编码效率。

papercode
论文代表图：DT-JRD: Deep Transformer based Just Recognizable Difference Prediction Model for Video Coding for Machines
DT-JRD: Deep Transformer based Just Recognizable Difference Prediction Model for Video Coding for Machines

Junqi Liu, Yun Zhang, Xiaoqi Wang, Long Xu, Sam Kwong. IEEE Transactions on Multimedia, vol. 28, pp. 114-127, 2026.

简介：将机器视觉能够辨认的最小差异定义为 JRD，并把预测过程建模为多类别分类任务。Transformer 模型联合提取内容与失真特征，使编码器在维持机器分析精度的同时减少不必要的码率。

papercode
论文代表图：Deep Learning based Joint Geometry and Attribute Upsampling for Large-Scale Colored Point Clouds
Deep Learning based Joint Geometry and Attribute Upsampling for Large-Scale Colored Point Clouds

Yun Zhang, Feifan Chen, Na Li, Xu Wang, Fen Miao, Sam Kwong. IEEE Transactions on Image Processing, vol. 35, pp. 1305-1320, 2026.

简介：提出联合几何与属性上采样框架 JGAU，利用二者的空间相关性恢复稠密点云。研究同时构建大规模彩色点云上采样数据，为不同放大倍率下的模型训练与评测提供支持。

papercode
论文代表图：C-CTX: Cubic-Checkerboard Context Entropy Model for Learned Image Compression
C-CTX: Cubic-Checkerboard Context Entropy Model for Learned Image Compression

Shiyu Feng, Linwei Zhu, Yun Zhang, Na Li, Wenhui Wu, Shiqi Wang. IEEE Transactions on Multimedia, vol. 28, pp. 1756-1766, 2026.

简介：提出立方棋盘上下文熵模型 C-CTX，在空间和通道维度统一组织与引用潜变量。通过通道重排和多种上下文预测方式，模型能够获得更准确的概率估计并改善学习式图像压缩性能。

papercode
论文代表图：VP-JND: Visual Perception Assisted Deep Picture-Wise Just Noticeable Distortion Prediction Model for Image Compression
VP-JND: Visual Perception Assisted Deep Picture-Wise Just Noticeable Distortion Prediction Model for Image Compression

Yun Zhang, Shisheng Zhang, Na Li, Chunling Fan, Raouf Hamzaoui. IEEE Transactions on Circuits and Systems for Video Technology, vol. 36, no. 1, pp. 622-636, 2026.

简介：提出图像级 VP-JND 预测模型，估计人眼刚好能够察觉的失真阈值。预测结果可用于指导编码参数和码率分配，在减少视觉冗余的同时尽量保持主观质量。

papercode
论文代表图：RegR-PCQA: Deep Learning based Colored Point Cloud Quality Assessment Using 3D-to-2D Regularized Representation
RegR-PCQA: Deep Learning based Colored Point Cloud Quality Assessment Using 3D-to-2D Regularized Representation

Yun Zhang, Cui Mao, Na Li, Chunling Fan, Weisi Lin. IEEE Transactions on Multimedia, vol. 28, pp. 1894-1908, 2026.

简介：将彩色点云规则化映射到二维几何图和属性图，再分别提取互补特征。框架结合 ViT 几何建模、CNN 属性建模与质量回归，实现面向彩色点云的客观质量评价。

papercode
论文代表图：Temporal Consistency-Aware Dynamic Point Clouds Color Attribute Enhancement
Temporal Consistency-Aware Dynamic Point Clouds Color Attribute Enhancement

Linwei Zhu, Ruxu Liang, Yun Zhang, Hui Yuan, Sam Kwong. IEEE Transactions on Multimedia, 2026.

简介：针对动态点云颜色属性在连续帧中的噪声与闪烁，设计三维时空搜索和时序特征融合流程。方法结合局部特征提取与 Conv-PointLSTM，利用帧间一致性恢复更稳定的颜色信息。

papercode

2025

论文代表图：Hierarchical Artifact Removal for Encoded Point Clouds with Very Low Bitrate G-PCC Octree
Hierarchical Artifact Removal for Encoded Point Clouds with Very Low Bitrate G-PCC Octree

Renwei Tu, Gangyi Jiang, Zhongjie Zhu, Yeyao Chen, Ting Luo, Yun Zhang, Mei Yu. IEEE Transactions on Emerging Topics in Computational Intelligence, vol. 10, no. 1, 2025.

简介：面向极低码率 G-PCC 八叉树编码造成的层级结构伪影，提出分层点云修复方法。模型在不同尺度上定位并校正失真区域，以改善稀疏编码条件下的几何结构与视觉质量。

papercode
论文代表图：Enhancing 3D Video Watching Experiences: Tackling Compression and 3D Warping Distortions in Synthesized View with Perceptual Guidance
Enhancing 3D Video Watching Experiences: Tackling Compression and 3D Warping Distortions in Synthesized View with Perceptual Guidance

Huan Zhang, Xu Zhang, Linwei Zhu, Yun Zhang, Jiangzhong Cao, Wing-Kuen Ling. Expert Systems With Applications, vol. 264, art. no. 125853, 2025.

简介：针对三维视频中同时存在的压缩失真和视图合成扭曲，构建感知引导的图像恢复网络。失真图预测分支与恢复分支协同工作，使模型能够有针对性地修复影响观看体验的区域。

papercode
论文代表图：Joint Multi-Dimensional Dynamic Attention and Transformer for General Image Restoration
Joint Multi-Dimensional Dynamic Attention and Transformer for General Image Restoration

Huan Zhang, Xu Zhang, Nian Cai, Jianglei Di, Yun Zhang. Computer Vision and Image Understanding, vol. 261, art. no. 104491, 2025.

简介：提出融合 CNN 与 Transformer 的 MDDA-former，并以多维动态注意力同时建模局部纹理和长程依赖。统一的编码器-解码器结构可适配去噪、去雨、去模糊等多类通用图像恢复任务。

papercode
论文代表图：LFIC-DRASC: Deep Light Field Image Compression Using Disentangled Representations and Asymmetrical Strip Convolution
LFIC-DRASC: Deep Light Field Image Compression Using Disentangled Representations and Asymmetrical Strip Convolution

Shiyu Feng, Yun Zhang, Linwei Zhu, Sam Kwong. IEEE Transactions on Broadcasting, vol. 71, no. 3, pp. 889-902, 2025.

简介：针对光场图像的高维结构和跨视角冗余，提出 LFIC-DRASC 端到端压缩框架。模型利用解耦表示、非对称条带卷积和专门的熵模型，在保持视角一致性的同时提升压缩效率。

papercode
论文代表图：Multi-Granular Embedding Optimization with Spatial-Channel Adaptive Tuning for Perceptual Image Quality Assessment
Multi-Granular Embedding Optimization with Spatial-Channel Adaptive Tuning for Perceptual Image Quality Assessment

Xiaoqi Wang, Yun Zhang, Junqi Liu. Neurocomputing, vol. 650, art. no. 130731, 2025.

简介：从空间和通道两个维度构造多粒度辅助嵌入，并通过自适应调节增强感知质量特征。框架进一步利用搜索策略选择合适的融合位置与顺序，从而提升跨数据集和跨失真类型的泛化能力。

papercode
论文代表图：Multi-scale Feature Importance-based Bit Allocation for End-to-End Feature Coding for Machines
Multi-scale Feature Importance-based Bit Allocation for End-to-End Feature Coding for Machines

Junle Liu, Yun Zhang, Zixi Guo, Xiaoxia Huang, Gangyi Jiang. ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 21, no. 9, art. no. 263, 2025.

简介：提出面向机器视觉特征编码的多尺度比特分配方法 MFIBA。方法估计不同尺度特征的重要性，并结合任务损失-码率模型进行在线分配，在压缩开销与下游任务性能之间取得平衡。

papercode
论文代表图：Texture-Aware Fast Mode Decision and Complexity Allocation for VVC Based Point Cloud Compression
Texture-Aware Fast Mode Decision and Complexity Allocation for VVC Based Point Cloud Compression

Lewen Fan, Yun Zhang. Journal of Visual Communication and Image Representation, vol. 113, art. no. 104610, 2025.

简介：针对 VVC 点云投影视频编码的高计算复杂度，提出纹理感知的快速模式决策与复杂度分配框架。模型利用纹理和编码单元相关性提前筛选模式，并根据内容特征动态分配计算资源。

papercode
论文代表图：TSC-PCAC: Voxel Transformer and Sparse Convolution-Based Point Cloud Attribute Compression for 3D Broadcasting
TSC-PCAC: Voxel Transformer and Sparse Convolution-Based Point Cloud Attribute Compression for 3D Broadcasting

Zixi Guo, Yun Zhang, Linwei Zhu, Hanli Wang, Gangyi Jiang. IEEE Transactions on Broadcasting, vol. 71, no. 1, pp. 154-166, 2025.

简介：提出融合体素 Transformer 与稀疏卷积的点云属性压缩框架 TSC-PCAC。两阶段模块分别捕获局部邻域与长程依赖，并结合通道上下文模型提高彩色属性的熵编码效率。

papercode
论文代表图：Geometry-Guided Latent Diffusion Model for Static Point Cloud Color Attribute Denoising
Geometry-Guided Latent Diffusion Model for Static Point Cloud Color Attribute Denoising

Linwei Zhu, Ruxu Liang, Yun Zhang, Gangyi Jiang, Yo-Sung Ho. IEEE Signal Processing Letters, vol. 32, pp. 2947-2951, 2025.

简介：以点云几何作为条件，在潜空间中建立颜色属性扩散去噪模型。方法使用更贴近真实压缩失真的噪声模拟和动态采样步数控制，在保持结构一致性的同时兼顾恢复质量与计算效率。

papercode
论文代表图：Texture and Structural Distortion Metric Based on Dual-Tree Complex Wavelet Transform for DIBR-Synthesized Image Quality Assessment
Texture and Structural Distortion Metric Based on Dual-Tree Complex Wavelet Transform for DIBR-Synthesized Image Quality Assessment

Huan Zhang, Zhijun Xiong, Xu Zhang, Jiangzhong Cao, Yun Zhang. Digital Signal Processing, vol. 165, art. no. 105293, 2025.

简介：利用 ResNet-50 与双树复小波变换提取多尺度、多方向的纹理和结构特征。模型结合全局物体位移补偿、方向失真分布和加权质量回归，对 DIBR 合成图像进行更有针对性的质量评价。

papercode
论文代表图：RGB-D Data Compression via Bi-Directional Cross-Modal Prior Transfer and Enhanced Entropy Modeling
RGB-D Data Compression via Bi-Directional Cross-Modal Prior Transfer and Enhanced Entropy Modeling

Yuyu Xu, Pingping Zhang, Minghui Chen, Zhuo Chen, Wenhui Wu, Yun Zhang, Xu Wang. ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 21, no. 2, art. no. 58, 2025.

简介：提出 RGB 与深度数据的联合压缩框架，在编码器、解码器和熵模型中建立双向跨模态连接。Bi-CPT 与 Bi-CEE 模块充分交换颜色和深度先验，并通过联合训练提高两种数据的率失真性能。

papercode
论文代表图：Influence of Inner-Core Symmetry on Tropical Cyclone Rapid Intensification and Its Forecasting by a Machine Learning Ensemble Model
Influence of Inner-Core Symmetry on Tropical Cyclone Rapid Intensification and Its Forecasting by a Machine Learning Ensemble Model

Jiali Zhang, Qinglan Li, Liguang Wu, Qifeng Qian, Xuyang Ge, Sam Kwong, Yun Zhang, Xinyan Lyu, Guanbo Zhou, Gaozhen Nie, Pak Wai Chan, Wai Kin Wong, Linwei Zhu. Weather and Climate Extremes, vol. 48, art. no. 100770, 2025.

简介：研究热带气旋内核对称性与快速增强之间的关系，并构建多种树模型及集成预测方案。模型融合卫星图像、强度和环境变量，为不同预报时效下的快速增强事件提供客观判断。

papercode

2024

论文代表图：Colored Point Cloud Quality Assessment Using Complementary Features in 3D and 2D Spaces
Colored Point Cloud Quality Assessment Using Complementary Features in 3D and 2D Spaces

Mao Cui, Yun Zhang, Chunling Fan, Raouf Hamzaoui, Qinglan Li. IEEE Transactions on Multimedia, vol. 26, pp. 11111-11125, 2024.

简介：提出 CF-PCQA，将三维点域的局部结构与二维投影域的纹理信息作为互补特征。多分支特征提取与回归模块共同学习几何、颜色和空间分布对主观质量的影响。

papercode
论文代表图：Learning to Predict Object-Wise Just Recognizable Distortion for Image and Video Compression
Learning to Predict Object-Wise Just Recognizable Distortion for Image and Video Compression

Yun Zhang, Haoqing Lin, Jing Sun, Linwei Zhu, Sam Kwong. IEEE Transactions on Multimedia, vol. 26, pp. 5925-5938, 2024.

简介：把恰可识别失真从图像级细化到物体级，并建立相应的数据生成与预测流程。模型可针对不同目标区域给出更精细的感知阈值，为图像和视频内容自适应压缩提供依据。

papercode
论文代表图：Neural Network Based Multi-Level In-Loop Filtering for Versatile Video Coding
Neural Network Based Multi-Level In-Loop Filtering for Versatile Video Coding

Linwei Zhu, Yun Zhang, Na Li, Wenhui Wu, Shiqi Wang, Sam Kwong. IEEE Transactions on Circuits and Systems for Video Technology, vol. 34, no. 11, pp. 12092-12096, 2024.

简介：在 VVC 编码环路中引入参考像素级、编码单元级和帧级三种神经滤波模块。不同层级分别处理局部预测误差、块级失真和全局伪影，联合使用可进一步改善重建质量与编码效率。

papercode
论文代表图：Quality Assessment for DIBR-Synthesized Views Based on Wavelet Transform and Gradient Magnitude Similarity
Quality Assessment for DIBR-Synthesized Views Based on Wavelet Transform and Gradient Magnitude Similarity

Huan Zhang, Dongsheng Zheng, Yun Zhang, Jianzhong Cao, Weisi Lin, Wing-Kuen Ling. IEEE Transactions on Multimedia, vol. 26, pp. 6834-6847, 2024.

简介：针对 DIBR 合成视图中的黑洞、拉伸和结构错位，结合小波分解与梯度幅值相似性提取失真特征。方法通过参考视图补偿和多尺度质量融合，提高对局部几何失真的感知敏感度。

papercode
论文代表图：Dynamic Weighted Gradient Reversal Network for Visible-Infrared Person Re-Identification
Dynamic Weighted Gradient Reversal Network for Visible-Infrared Person Re-Identification

Chenghua Li, Zongze Li, Jing Sun, Yun Zhang, Xiaoping Jiang, Fan Zhang. ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 20, no. 1, art. no. 12, 2024.

简介：提出用于可见光-红外行人重识别的动态加权梯度反转网络 DGRNet。网络联合身份分类与模态判别，并根据训练状态调节对抗强度，以缩小模态差异并保留身份特征。

papercode
论文代表图：Multiview Projection Based Joint Geometry and Color Hole Repairing Method for G-PCC Trisoup Encoded Color Point Cloud
Multiview Projection Based Joint Geometry and Color Hole Repairing Method for G-PCC Trisoup Encoded Color Point Cloud

Taowen Xu, Gangyi Jiang, Mei Yu, Yun Zhang, Zhidi Jiang, Yo-Sung Ho. IEEE Transactions on Emerging Topics in Computational Intelligence, vol. 8, no. 1, pp. 892-902, 2024.

简介：针对 G-PCC Trisoup 编码产生的几何孔洞和颜色缺失，提出多视图投影联合修复框架。几何加权修复与门控颜色修复网络协同工作，最后通过逆投影重建更完整的彩色点云。

papercode
论文代表图：PCQD-AR: Subjective Quality Assessment of Compressed Point Clouds with Head-Mounted Augmented Reality
PCQD-AR: Subjective Quality Assessment of Compressed Point Clouds with Head-Mounted Augmented Reality

Chunling Fan, Yun Zhang, Linwei Zhu, Xinju Wu. IET Electronics Letters, vol. 60, no. 5, pp. 1-4, 2024.

简介：在头戴式增强现实环境中开展压缩点云主观质量实验，并构建包含 10 个参考模型和 90 个失真版本的数据集。研究分析几何和纹理量化对六自由度观看体验的影响，为客观评价指标提供基准。

papercode
论文代表图：Cracks-Suppression Perceptual Geometry Coding for Dynamic Point Clouds
Cracks-Suppression Perceptual Geometry Coding for Dynamic Point Clouds

Wei Liu, Mei Yu, Zhidi Jiang, Haiyong Xu, Zhoujie Zhu, Yun Zhang, Gangyi Jiang. Digital Signal Processing, vol. 149, pp. 1-13, 2024.

简介：针对 V-PCC 投影、填充与量化过程中容易出现的裂纹伪影，提出面向动态点云的感知几何编码方法。模型改进几何图生成与压缩控制，使关键结构区域在重建后保持更好的连续性。

papercode
论文代表图：Optimized Quantization Parameter Selection for Video-Based Point Cloud Compression
Optimized Quantization Parameter Selection for Video-Based Point Cloud Compression

Hui Yuan, Raouf Hamzaoui, Ferrante Neri, Shengxiang Yang, Xin Lu, Linwei Zhu, Yun Zhang. Frontiers in Signal Processing, vol. 4, art. no. 1385287, 2024.

简介：将率失真优化建模用于 V-PCC 量化参数选择，并采用差分进化搜索不同帧组的参数组合。该方法根据内容和码率目标自适应配置量化强度，从而降低码率并保持点云重建质量。

papercode
论文代表图：Guest Editorial: Deep Learning-Based Point Cloud Processing, Compression and Analysis
Guest Editorial: Deep Learning-Based Point Cloud Processing, Compression and Analysis

Yun Zhang, Raouf Hamzaoui, Xu Wang, Junhui Hou, Giuseppe Valenzise. IET Electronics Letters, vol. 60, no. 14, pp. 1-2, 2024.

简介：本专题社论梳理深度学习在点云处理、压缩与分析中的最新研究进展。内容涵盖采样、增强、编码、质量评价和三维理解等方向，并概括专题论文的主要技术贡献。

papercode

论文

已发表 SCI/EI 学术论文 180 余篇，其中 IEEE/ACM Transactions 70 余篇。

专著

出版学术专著《三维视频处理》一部。

专利

授权中/美/PCT 发明专利 50 余项。