Compositing is one of the most important editing operations for images and videos. The process of improving the realism of composite results is often called harmonization. Previous approaches for harmonization mainly focus on images. In this work, we take one step further to attack the problem of video harmonization. Specifically, we train a convolutional neural network in an adversarial way, exploiting a pixel-wise disharmony discriminator to achieve more realistic harmonized results and introducing a temporal loss to increase temporal consistency between consecutive harmonized frames. Thanks to the pixel-wise disharmony discriminator, we are also able to relieve the need of input foreground masks. Since existing video datasets which have ground-truth foreground masks and optical flows are not sufficiently large, we propose a simple yet efficient method to build up a synthetic dataset supporting supervised training of the proposed adversarial network. Experiments show that training on our synthetic dataset generalizes well to the real-world composite dataset. Also, our method successfully incorporates temporal consistency during training and achieves more harmonious results than previous methods.
Segmentation-guided Domain Adaptation and Data Harmonization of Multi-device Retinal Optical Coherence Tomography using Cycle-Consistent Generative Adversarial Networks
Simon Fraser University（西蒙弗雷泽大学）；Wake Forest University School of Medicine（维克森林大学医学院）；Wake Forest School of Medicine（维克森林大学医学院）；University of Nottingham（诺丁汉大学）；Tencent Jarvis Lab（腾讯Jarvis实验室）；Memorial University of Newfoundland（纽芬兰纪念大学）；University of British Columbia（不列颠哥伦比亚大学）；The University of British Columbia（不列颠哥伦比亚大学）；University College London（伦敦大学学院）
FedSaaS: Class-Consistency Federated Semantic Segmentation via Global Prototype Supervision and Local Adversarial Harmonization
Shandong Computer Science Center, Qilu University of Technology (Shandong Academy of Sciences)（山东省计算中心（齐鲁工业大学（山东省科学院）））；Shandong Fundamental Research Center for Computer Science（山东省计算机科学基础研究中心）；College of Control Science and Engineering, Zhejiang University（浙江大学控制科学与工程学院）
Dynamic Kernel Distillation for Efficient Pose Estimation in Videos
Department of Electrical and Computer Engineering, National University of Singapore, Singapore（新加坡国立大学电气与计算机工程系）；Snap Inc.（Snap公司）；ByteDance AI Lab（字节跳动人工智能实验室）
Segmentation-Renormalized Deep Feature Modulation for Unpaired Image Harmonization
New York University Tandon School of Engineering（纽约大学坦登工程学院）
深度网络现在在大规模多中心成像研究中无处不在。然而，由于对比度、分辨率和噪声不一致，跨站点的图像直接聚合禁忌用于下游统计和基于深度学习的图像分析。为此，在没有配对数据的情况下，周期一致生成对抗网络的变体已被用于协调源域和目标域之间的图像集。重要的是，这些方法易于不稳定，对比度反转，棘手的病理操纵，和隐写映射，限制了它们在现实世界的医学成像中的可靠采用。在这项工作中，基于一个基本的假设，即形态形状是一致的成像部位，我们提出了一个分割重新规范化的图像翻译框架，以减少扫描仪间的异质性，同时保留解剖布局。我们用以共同学习的解剖分割嵌入为条件的可训练尺度和移位参数替换生成网络内标准化层中使用的仿射变换，以调节每个翻译级别的特征。我们在有和没有病变的数据集上，针对几种成像模式（T1 w MRI、FLAIR MRI和OCT）的最新基线评估了我们的方法。平移GANs的分割重正化产生了由Inception距离量化的卓越图像协调性，通过事后分割准确性证明了改进的下游效用，并提高了对平移扰动和自对抗攻击的鲁棒性。
Latency-aware Road Anomaly Segmentation in Videos: A Photorealistic Dataset and New Metrics
AIR, Tsinghua University（清华大学智能产业研究院）；Beijing Institute of Technology（北京理工大学）；Chinese Academy of Science（中国科学院）；Hong Kong Polytechnic University（香港理工大学）；Mercedes-Benz Group China Ltd.（梅赛德斯-奔驰中国有限公司）
Acquisition of interpretable domain information during brain MR image harmonization for content-based image retrieval
Dept. of Applied Informatics, Graduate School of Science and Engineering, Hosei University（应用信息科，理学部，法政大学）；Dept. of Radiology and Radiological Science, Johns Hopkins Medicine（放射科和放射科学系，约翰霍普金斯医学中心）
Harmonization Across Imaging Locations(HAIL): One-Shot Learning for Brain MRI
Children’s National Hospital（美国儿童国家医疗中心）；George Washington University（乔治·华盛顿大学）；Children’s Hospital Colorado（科罗拉多儿童医院）；Children’s Hospital of Philadelphia（费城儿童医院）
Inter-vendor harmonization of Computed Tomography (CT) reconstruction kernels using unpaired image translation
Vanderbilt University（范德堡大学）；Vanderbilt University Medical Center（范德堡大学医学中心）；Amsterdam University Medical Center（阿姆斯特丹大学医学中心）；University of Amsterdam（阿姆斯特丹大学）
Investigating the impact of kernel harmonization and deformable registration on inspiratory and expiratory chest CT images for people with COPD
Vanderbilt University（范德堡大学）；Insitro；Vanderbilt University Medical Center（范德堡大学医学中心）；Department of Veterans Affairs Medical Center（退伍军人事务部医疗中心）；Vanderbilt University Institute of Imaging Science（范德堡大学成像科学研究所）
针对 COPD 患者吸呼气 CT 因重建核差异导致定量分析误差的问题，提出两阶段流程：先用循环 GAN 进行核 harmonization，再进行可变形配准，显著降低肺气肿测量不一致性并提高配准精度。
T2VAttack: Adversarial Attack on Text-to-Video Diffusion Models
Hangzhou Institute for Advanced Study, UCAS, school of Intelligent Science and Technology（杭州高等研究院，UCAS，智能科学与技术学院）；State Key Laboratory of AI Safety, Institute of Computing Technology (ICT), Chinese Academy of Sciences (CAS)（人工智能安全国家重点实验室，计算技术研究所，中国科学院）；University of Chinese Academy of Sciences (UCAS)（中国科学院大学）
TAI-GAN: A Temporally and Anatomically Informed Generative Adversarial Network for early-to-late frame conversion in dynamic cardiac PET inter-frame motion correction
Yale University（耶鲁大学）；IBM Research（IBM研究院）；Yale New Haven Hospital（耶鲁-纽黑文医院）；Yale School of Medicine（耶鲁医学院）
ReToMe-VA: Recursive Token Merging for Video Diffusion-based Unrestricted Adversarial Attack
School of Computer Science, Fudan University（复旦大学计算机科学技术学院）；Artificial Intelligence Innovation and Incubation Institute, Fudan University（复旦大学人工智能创新与孵化研究院）
Tsinghua National Lab for Information Science and Technology（清华大学信息科学与技术国家实验室）；State Key Lab of Intelligent Technology and Systems（智能技术与系统国家重点实验室）；Center for Bio-Inspired Computing Research（生物启发计算研究中心）；Department of Computer Science and Technology, Tsinghua University（清华大学计算机科学与技术系）
VideoPure: Diffusion-based Adversarial Purification for Video Recognition
Shanghai Engineering Research Center of AI Robotics, Academy for Engineering & Technology, Fudan University（复旦大学工程与技术学院上海市人工智能机器人工程研究中心）；Engineering Research Center of AI & Robotics, Ministry of Education, Academy for Engineering & Technology, Fudan University（复旦大学工程与技术学院教育部人工智能与机器人工程研究中心）；School of Computer Science, Fudan University（复旦大学计算机科学学院）
Dynamically Masked Discriminator for Generative Adversarial Networks
King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）；National University of Singapore（新加坡国立大学）；Tencent YouTu Lab（腾讯优图实验室）；Guangxi Medical University（广西医科大学）
