A flow model for joint action recognition and identity maintenance

机译：联合行动识别和身份维护的流模型

获取原文

获取原文并翻译 | 示例

页面导航

摘要
著录项
相似文献
相关主题

摘要

We propose a framework that performs action recognition and identity maintenance of multiple targets simultaneously. Instead of first establishing tracks using an appearance model and then performing action recognition, we construct a network flow-based model that links detected bounding boxes across video frames while inferring activities, thus integrating identity maintenance and action recognition. Inference in our model reduces to a constrained minimum cost flow problem, which we solve exactly and efficiently. By leveraging both appearance similarity and action transition likelihoods, our model improves on state-of-the-art results on action recognition for two datasets.

机译：我们提出了一个框架，可以同时执行动作识别和多个目标的身份维护。与其先使用外观模型建立轨迹，然后执行动作识别，不如构造一个基于网络流的模型，该模型将跨视频帧检测到的边界框链接在一起，同时推断活动，从而集成了身份维护和动作识别。我们模型中的推论减少到一个受约束的最小成本流问题，我们可以准确而有效地解决该问题。通过利用外观相似性和动作转换可能性，我们的模型改进了两个数据集动作识别的最新结果。

著录项

来源
《Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on》|2012年|p.1218- 1225|共8页
会议地点 Providence RI(US)
作者
Khamis Sameh; Morariu Vlad I.; Davis Larry S.;
展开▼
作者单位

University of Maryland, College Park;

展开▼
会议组织
原文格式 PDF
正文语种 eng
中图分类 TP391.41;
关键词

相似文献

外文文献
中文文献
专利

1. Scientific Opinion on the substantiation of health claims related to gamma linolenic acid and maintenance of joints (ID 494, 637, 1774, 2098), weight maintenance following weight loss (ID 496), maintenance of peripheral blood flow (ID 638), maintenance of normal blood pressure (ID 1771), maintenance of normal blood cholesterol concentrations (ID 1771) and maintenance of bone (ID 1774) pursuant to Article 13(1) of Regulation (EC) No 1924/2006 [J] . EFSA Journal . 2010,第2期

机译：关于证实与γ-亚麻酸有关的健康主张和维持关节的科学意见（ID 494、637、1774、2098），减肥后的体重维持（ID 496），维持外周血流（ID 638），维持根据（EC）第1924/2006号法规第13（1）条，维持正常血压（ID 1771），维持正常胆固醇水平（ID 1771）和维持骨骼（ID 1774）
2. Multi-perspective and multi-modality joint representation and recognition model for 3D action recognition [J] . Gao Z., Zhang H., Xu G. P., Neurocomputing . 2015,第mara5pta2期

机译：用于3D动作识别的多视角，多模态联合表示和识别模型
3. A discriminative structural model for joint segmentation and recognition of human actions [J] . Liu Cuiwei, Hou Jingyi, Wu Xinxiao, Multimedia Tools and Applications . 2018,第24期

机译：区分和识别人类行为的区别性结构模型
4. A flow model for joint action recognition and identity maintenance [C] . Khamis Sameh, Morariu Vlad I., Davis Larry S. IEEE Conference on Computer Vision and Pattern Recognition . 2012

机译：联合动作识别与身份维护的流模型
5. Facial expression recognition using Optical Flow and 3D HMM and human action recognition using cuboid and topic models [D] . Kung, Sheng H. 2016

机译：使用Optical Flow和3D HMM进行面部表情识别，并使用长方体和主题模型进行人体动作识别
6. RNA recognition by the joint action of two nucleolin RNA-binding domains: genetic analysis and structural modeling. [O] . P Bouvet, C Jain, J G Belasco, 1997

机译：通过两个核仁素RNA结合域的联合作用进行RNA识别：遗传分析和结构建模。
7. A flow model for joint action recognition and identity maintenance [O] . Sameh Khamis, Vlad I. Morariu, Larry S. Davis 2012

机译：用于联合行动识别和身份维护的流程模型
8. Coupled Mechanical-Hydrological Methodology for Modeling Flow in Jointed Rock Masses Using Laboratory Data for the Joint Flow Model [R] . Voss, C. F., Bastian, R. J., Shotwell, L. R. 1986

机译：利用实验室数据建立节理岩体流动的联合流动模型耦合机械水文方法

A flow model for joint action recognition and identity maintenance

摘要

著录项

相似文献

相关主题

期刊订阅