基于随机丛林的人体部件分类方法

计算机工程

基于随机丛林的人体部件分类方法

屈雁秋,陈锋

(中国科学技术大学自动化系,合肥 230000)

收稿日期:2017-01-22 出版日期:2018-01-15 发布日期:2018-01-15
作者简介:屈雁秋(1992—),男,硕士研究生,主研方向为计算机视觉、机器学习;陈锋,副教授、博士。

Body Part Classification Method Based on Random Jungle

QU Yanqiu,CHEN Feng

(Department of Automation,University of Science and Technology of China,Hefei 230000,China)

Received:2017-01-22 Online:2018-01-15 Published:2018-01-15

摘要/Abstract

摘要： 人体部件分类是人体姿态跟踪中的重要前提和关键步骤。传统深度梯度特征下的随机森林分类方法虽然可以达到实时性的要求,但仍存在准确度不高、对噪声不够鲁棒、内存消耗过大等缺点。为此,提出传统深度特征与RGB边缘特征相融合的一种新的分类方法,并在特征维度加大的情况下引入随机丛林模型。实验结果表明,该特征分类方法不仅可以减少20%左右的运行时间,还可以提高1%左右的测试准确率。

关键词: 人体部件分类, 深度梯度特征, RGB边缘特征, 随机丛林, 有向无环图

Abstract: Body part classification is an important precondition and crucial step in human pose tracking.Although the random forest classical method with gradient feature of depth can get the request of real time,it remains the weakness of low accuracy,not robust enough to the noise and huge memory consumption.This paper proposes a new classification method,which combines the classical depth feature with RGB edge feature,and introduces the random jungle model to solve the increased dimension of the feature.Experimental results show that the proposed method not only saves about 20% running time,but also improves the test accuracy of about 1%.

Key words: body part classification, gradient feature of depth, edge feature of RGB, random jungle, Directed Acyclic Graph(DAG)

中图分类号:

TP391.41

屈雁秋,陈锋. 基于随机丛林的人体部件分类方法[J]. 计算机工程.

QU Yanqiu,CHEN Feng. Body Part Classification Method Based on Random Jungle[J]. Computer Engineering.

参考文献

参考文献［1］DANTONE M,GALL J,LEISTNER C,et al.Human Pose Estimation Using Body Parts Dependent Joint Reg-ressors［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2013:3041-3048. ［2］GIRSHICK R,SHOTTON J,KOHLI P,et al.Efficient Regression of General-activity Human Poses from Depth Images［C］//Proceedings of International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2011:415-422. ［3］YE M,WANG X,YANG R,et al.Accurate 3D Pose Estimation from a Single Depth Image［C］//Proceedings of International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2011:731-738. ［4］汪成峰,王庆,梅树立,等.基于插值小波关键帧提取的动作评价算法［J］.计算机工程,2017,43(1):309-315. ［5］BREIMAN L.Random Forests［J］.Machine Learning,2001,45(1):5-32. ［6］SHOTTON J,FITZGIBBON A,COOK M,et al.Real-time Human Pose Recognition in Parts from Single Depth Images［J］.Communications of the ACM,2013,56(1):116-124. ［7］SHOTTON J,NOWOZIN S,SHARP T,et al.Decision Jungles:Compact and Rich Models for Classification［C］//Proceedings of International Conference on Neural Information Processing Systems.New York,USA:ACM Press,2013:234-242. ［8］LEPETIT V,LAGGER P,FUA P.Randomized Trees for Real-time Keypoint Recognition［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2005:775-781. ［9］郭天楚,吴晓雨,杨磊,等.基于改进的随机森林的人体部件识别［J］.中国传媒大学学报(自然科学版),2014,21(5):32-38. ［10］徐岳峰,周书仁,王刚,等.基于深度图像梯度特征的人体姿态估计［J］.计算机工程,2015,41(12):200-205. ［11］JIU Mingyuan,WOLF C,BASKURT A.Integrating Spatial Layout of Object Parts into Classification Without Pairwise Terms——Application to Fast Body Parts Estimation from Depth Images［C］//Proceedings of International Joint Conference on Computer Vision,Imaging and Computer Graphics Theory and Applications.Berlin,Germany:Springer,2013. ［12］LEPETIT V,PILET J,FUA P.Point Matching as a Classification Problem for Fast and Robust Object Pose Estimation［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2004:244-250. ［13］WU Jianxin,GEYER C,REHG J M.Real-time Human Detection Using Contour Cues［C］//Proceedings of IEEE International Conference on Robotics & Automation.Washington D.C.,USA:IEEE Press,2011:860-867. ［14］ZHANG Huigang,WANG Junxiu,BAI Xiao,et al.Object Detection via Foreground Contour Feature Selection and Part-based Shape Model［C］//Proceedings of the 21st International Conference on Pattern Recognition.Washington D.C.,USA:IEEE Press,2012:2524-2527. ［15］SHOTTON J,BLAKE A,CIPOLLA R.Multiscale Categorical Object Recognition Using Contour Fragments［J］.IEEE Transactions on Pattern Analysis & Machine Intelligence,2008,30(7):1270-81. ［16］BARROW H G,TENENBAUM J M,BOLLES R C,et al.Parametric Correspondence and Chamfer Matching:Two New Techniques for Image Matching［C］//Proceedings of the 5th International Joint Conference on Artificial Intelligence.New York,USA:ACM Press,1977:659-663. 编辑顾逸斐

[1]	刘寅昊, 蒋文保, 孙林昆, 王勇攀. 基于路径存储表的Hashgraph共识算法优化与实现[J]. 计算机工程, 2024, 50(6): 166-178.
[2]	王劲松, 杨唯正, 赵泽宁, 魏佳佳. 基于有向无环图的区块链技术综述[J]. 计算机工程, 2022, 48(6): 11-23.
[3]	王静, 陈岚, 张贺, 王海永. 基于EDA多任务流的调度算法研究[J]. 计算机工程, 2021, 47(2): 146-151.
[4]	郭彪, 唐麒, 文智敏, 傅娟, 王玲, 魏急波. 面向SCA的DPR软件架构设计与调度技术[J]. 计算机工程, 2021, 47(12): 221-229.
[5]	焦一鸣, 周川, 郭健, 崔玉伟. 异构计算环境下一种新型的多DAG任务调度算法[J]. 计算机工程, 2019, 45(7): 1-5.
[6]	李云洋,周川,王琦. 异构分布式计算环境下一种新型表调度算法[J]. 计算机工程, 2018, 44(8): 43-47.
[7]	杨平平,岳春生,胡泽明. 异构信号处理平台中层次性流水线调度算法[J]. 计算机工程, 2018, 44(11): 83-89.
[8]	沈健,蒋芸,邹丽,陈娜,胡学伟. 基于节点选择优化的DAG-SVM 多类别分类[J]. 计算机工程, 2015, 41(6): 143-146.
[9]	刘宴涛,夏桂阳,徐静,秦娜. 一种基于子树分解的组播线性网络编码算法[J]. 计算机工程, 2015, 41(11): 153-159.
[10]	黄袁, 付晓东, 贾楠, 代志华, 马玉倩. 一种工作流的控制流距离度量方法[J]. 计算机工程, 2013, 39(7): 94-98,114.
[11]	叶晓敏, 王侃文, 陈佳临, 周学功, 王伶俐. 优化的可重构多常数乘法器生成算法[J]. 计算机工程, 2012, 38(3): 228-233.
[12]	刘石柱, 尹首一, 殷崇勇, 刘雷波, 魏少军. 基于可重构处理器的并行优化算法[J]. 计算机工程, 2012, 38(21): 286-289.
[13]	李静梅, 张博. 基于混合粒子群优化的CMP线程调度方法[J]. 计算机工程, 2012, 38(20): 113-115.
[14]	孙超, 于策, 孙济洲, 王萌萌, 杜君. 面向并行应用程序的DAG模式复用技术[J]. 计算机工程, 2012, 38(11): 66-68.
[15]	赵磊, 吕莉媛, 赵国生. 一种软件体系结构级基本路径覆盖方法[J]. 计算机工程, 2012, 38(10): 60-62.

选择文件类型/文献管理软件名称

选择包含的内容