Predicting human interaction is challenging as the on-going activity has to be inferred based on a partially observed video. Essentially, a good algorithm should effectively model the mutual influence between the two interacting subjects. Also, only a small region in the scene is discriminative for identifying the on-going interaction. In this work, we propose a relative attention model to explicitly address these difficulties. Built on a tri-coupled deep recurrent structure representing both interacting subjects and global interaction status, the proposed network collects spatio-temporal information from each subject, rectified with global interaction information, yielding effective interaction representation. Moreover, the proposed network also unifies an attention module to assign higher importance to the regions which are relevant to the on-going action. Extensive experiments have been conducted on two public datasets, and the results demonstrate that the proposed relative attention network successfully predicts informative regions between interacting subjects, which in turn yields superior human interaction prediction accuracy.
MagNet: Multi-Level Attention Graph Network for Predicting High-Resolution Spatial Transcriptomics
Vanderbilt University(范德堡大学);Weill Cornell Medicine(威尔康奈尔医学院);NVIDIA(英伟达公司);Vanderbilt University Medical Center(范德堡大学医学中心);UT MD Anderson Cancer Center(得克萨斯大学 MD 安德森癌症中心)
Cross-attention-based saliency inference for predicting cancer metastasis on whole slide images
Center for Biomedical Informatics, Wake Forest University School of Medicine(维克森林大学医学院生物医学信息学中心);Department of Pathology, Wake Forest University School of Medicine(维克森林大学医学院病理学系)
An Improved ResNet50 Model for Predicting Pavement Condition Index (PCI) Directly from Pavement Images
TriQuint Semiconductor(TriQuint半导体公司);XLIM Laboratory(XLIM实验室);University of Limoges(利摩日大学);Department of Electrical, Computer and Energy Engineering(电气、计算机与能源工程系);University of Colorado(科罗拉多大学);Qualcomm Inc.(高通公司)
DeepRA: Predicting Joint Damage From Radiographs Using CNN with Attention
ZS Associates(ZS咨询公司)
风湿性关节炎(RA)的关节损伤通过手动检查和分级手和脚的X光片来评估。这是一项繁琐的任务,需要训练有素的专家,他们的主观评估导致评分者之间的一致性很低。一种可以自动预测手脚关节水平损伤的算法可以帮助优化这一过程,这最终将有助于医生更好地护理和研究患者。在本文中,我们提出了一个两阶段的方法,它融合了目标检测和卷积神经网络与注意力,可以有效地和准确地预测整体和关节水平的狭窄和侵蚀从患者的X光片。这种方法已经在患有RA的患者的手和脚X光片上进行了评估,并且在预测关节水平狭窄和侵蚀Sharp van der Heijde(SvH)评分方面实现了1.358和1.404的加权均方根误差(RMSE),相对于基线SvH评分分别改善了31%和19%。所提出的方法在预测患者手部和足部X光片的总体损伤方面实现了1.456的加权绝对误差,与基线相比改善了79%。我们的方法还提供了一种内在的能力,可以使用注意力权重来解释模型预测,这对于深度学习模型的黑盒性质是必不可少的。提出的方法是在由Dream Challenges主办的RA2 Dream Challenge期间开发的,并在预测X线片的整体和关节水平狭窄和侵蚀SvH评分方面获得了第4和第8名。
Measuring and Predicting Where and When Pathologists Focus their Visual Attention while Grading Whole Slide Images of Cancer
Department of Computer Science(计算机科学系);Department of Biomedical Informatics(生物医学信息学系);Department of Psychology(心理学系);Department of Pathology and Laboratory Medicine(病理学与实验室医学系);Department of Pathology(病理学系)
Ordinal Learning: Longitudinal Attention Alignment Model for Predicting Time to Future Breast Cancer Events from Mammograms
Netherlands Cancer Institute (NKI)(荷兰癌症研究所);GROW School for Oncology and Development Biology, Maastricht University(马斯特里赫特大学GROW肿瘤学与发育生物学学院);Macao Polytechnic University(澳门理工大学);Radboud University Medical Centre(拉德堡德大学医学中心)
Predicting Video Slot Attention Queries from Random Slot-Feature Pairs
Department of Electrical Engineering and Automation, Aalto University(艾罗大学电气工程与自动化系);Gaoling School of Artificial Intelligence, Renmin University of China(中国人民大学研究生院人工智能学院);Department of Computer Science, Aalto University(艾罗大学计算机科学系);Center for Machine Vision and Signal Analysis, University of Oulu(奥卢大学机器视觉与信号分析中心)
Predicting Blastocyst Formation in IVF: Integrating DINOv2 and Attention-Based LSTM on Time-Lapse Embryo Images
Sustainable Digitalisation Research Center, Department of Computer Science and Media Technology(可持续数字化研究中心、计算机科学与媒体技术系);Sustainable Digitalisation Research Center, Department of Computer Science(可持续数字化研究中心、计算机科学系);School of Information Technology(信息科技学院);Department of Computer Science(计算机科学系)
An attention-based deep learning network for predicting Platinum resistance in ovarian cancer
College of Medicine and Biological Information Engineering, Northeastern University(东北大学医学与生物信息工程学院);Shengjing Hospital of China Medical University(中国医科大学附属盛京医院)
Can Physician Judgment Enhance Model Trustworthiness? A Case Study on Predicting Pathological Lymph Nodes in Rectal Cancer
National Cancer Center Research Institute(国立癌症中心研究所);RIKEN Center for Advanced Intelligence Project(理化学研究所先进智能项目中心);National Cancer Center Hospital(国立癌症中心医院);The University of Tokyo(东京大学)
A dual-task mutual learning framework for predicting post-thrombectomy cerebral hemorrhage
ShanghaiTech University(上海科技大学);Imperial College London(帝国理工学院);Westlake University School of Medicine(西湖大学医学院);Affiliated Hangzhou First People’s Hospital(杭州市第一人民医院);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Royal Brompton Hospital(皇家布朗普顿医院);King’s College London(伦敦国王学院);Shanghai United Imaging Intelligence Co., Ltd.(上海联影智能医疗科技有限公司);Shanghai Clinical Research and Trial Center(上海临床研究与试验中心)
Diffusion Attention Expert Model for Predicting and Semi-automatic Localizing STAS in Lung Cancer Histopathological Images
College of Computer Science and Electronic Engineering, Hunan University(湖南大学计算机科学与电子工程学院);Department of Pathology, The Second Xiangya Hospital, Central South University(中南大学湘雅医院病理科);Hunan Clinical Medical Research Center for Cancer Pathogenic Genes Testing and Diagnosis(湖南临床医学肿瘤基因检测与诊断研究中心);Department of Thoracic Surgery, The Second Xiangya Hospital, Central South University(中南大学湘雅医院胸外科);Department of pathology, Hunan Cancer Hospital, The Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University(湖南肿瘤医院病理科);Department of Pathology, The Third Xiangya Hospital, Central South University(中南大学湘雅第三医院病理科);Department of Pathology, First People's Hospital of Pingjiang County(平江县第一人民医院病理科);Department of Pathology, the First Affiliated Hospital, Hengyang Medical School, University of South China(南华大学衡阳医学院第一附属医院病理科);Department of Radiology, The Second Xiangya Hospital of Central South University(中南大学湘雅医院放射科);Department of Radiology, Xiangya Hospital, Central South University(中南大学湘雅医院放射科);Oncology Department and State Key Laboratory of Systems Medicine for Cancer of Shanghai Cancer Institute, Renji Hospital, School of Medicine, Shanghai Jiaotong University(上海癌症研究院肿瘤科及上海交通大学医学院系统医学重点实验室)
"Is this an example image?" -- Predicting the Relative Abstractness Level of Image and Text
Leibniz Information Centre for Science and Technology (TIB)(莱布尼茨科学技术信息中心(TIB));L3S Research Center, Leibniz Universität Hannover(莱布尼茨汉诺威大学L3S研究中心);Leibniz Universität Hannover(莱布尼茨汉诺威大学);L3S Research Center(L3S研究中心)