This paper presents a multi-face detection method for color images. The method is based on the assumption that faces are well separated from the background by skin color detection. These faces can be located by the pr...This paper presents a multi-face detection method for color images. The method is based on the assumption that faces are well separated from the background by skin color detection. These faces can be located by the proposed method which modifies the subtractive clustering. The modified clustering algorithm proposes a new definition of distance for multi-face detection, and its key parameters can be predetermined adaptively by statistical information of face objects in the image. Downsampling is employed to reduce the computation of clustering and speed up the process of the proposed method. The effectiveness of the proposed method is illustrated by three experiments.展开更多
人脸姿态变化复杂且对人脸识别性能影响明显,提出了一种融合LCCDN (LSTM and CNN based Cascade Deep Network)与增量聚类的多姿态人脸识别方法。采用LCCDN模型定位人脸关键点,利用长短时记忆网络(LSTM)的记忆功能寻找人脸各关键点在空...人脸姿态变化复杂且对人脸识别性能影响明显,提出了一种融合LCCDN (LSTM and CNN based Cascade Deep Network)与增量聚类的多姿态人脸识别方法。采用LCCDN模型定位人脸关键点,利用长短时记忆网络(LSTM)的记忆功能寻找人脸各关键点在空间上的全局上下文的依赖关系对人脸关键点初始化,并通过卷积神经网络模型,采用由粗到精的策略;定位人脸关键点;以人脸关键点作为人脸朝向描述子,同时为适应人脸姿态不断地动态更新,采用基于熵诱导度量机制的增量聚类方法,对头部姿态进行动态增量聚类,构建人脸姿态池。在此基础上,通过建立不同姿态的人脸识别分类模型实现多姿态人脸识别,在CAS-PEAL-R1、CFP和Multi-PIE三个数据集上的人脸识别准确率分别达到96.75%,96.50%,97.82%。通过与同类人脸识别方法的客观定量对比,实验结果表明所提方法有效、可行。展开更多
In this study, we address the problems encountered by incremental face clustering. Without the benefit of having observed the entire data distribution, incremental face clustering is more challenging than static datas...In this study, we address the problems encountered by incremental face clustering. Without the benefit of having observed the entire data distribution, incremental face clustering is more challenging than static dataset clustering. Conventional methods rely on the statistical information of previous clusters to improve the efficiency of incremental clustering;thus, error accumulation may occur. Therefore, this study proposes to predict the summaries of previous data directly from data distribution via supervised learning. Moreover, an efficient framework to cluster previous summaries with new data is explored. Although learning summaries from original data costs more than those from previous clusters, the entire framework consumes just a little bit more time because clustering current data and generating summaries for new data share most of the calculations. Experiments show that the proposed approach significantly outperforms the existing incremental face clustering methods, as evidenced by the improvement of average F-score from 0.644 to 0.762. Compared with state-of-the-art static face clustering methods, our method can yield comparable accuracy while consuming much less time.展开更多
文摘This paper presents a multi-face detection method for color images. The method is based on the assumption that faces are well separated from the background by skin color detection. These faces can be located by the proposed method which modifies the subtractive clustering. The modified clustering algorithm proposes a new definition of distance for multi-face detection, and its key parameters can be predetermined adaptively by statistical information of face objects in the image. Downsampling is employed to reduce the computation of clustering and speed up the process of the proposed method. The effectiveness of the proposed method is illustrated by three experiments.
文摘人脸姿态变化复杂且对人脸识别性能影响明显,提出了一种融合LCCDN (LSTM and CNN based Cascade Deep Network)与增量聚类的多姿态人脸识别方法。采用LCCDN模型定位人脸关键点,利用长短时记忆网络(LSTM)的记忆功能寻找人脸各关键点在空间上的全局上下文的依赖关系对人脸关键点初始化,并通过卷积神经网络模型,采用由粗到精的策略;定位人脸关键点;以人脸关键点作为人脸朝向描述子,同时为适应人脸姿态不断地动态更新,采用基于熵诱导度量机制的增量聚类方法,对头部姿态进行动态增量聚类,构建人脸姿态池。在此基础上,通过建立不同姿态的人脸识别分类模型实现多姿态人脸识别,在CAS-PEAL-R1、CFP和Multi-PIE三个数据集上的人脸识别准确率分别达到96.75%,96.50%,97.82%。通过与同类人脸识别方法的客观定量对比,实验结果表明所提方法有效、可行。
基金supported by the National Natural Science Foundation of China (Nos. 61701277 and 61771288)the State Key Development Program in13th Five-Year (Nos. 2016YFB0801301, 044007008, and 2016YFB1001005)supported by the National Engineering Laboratory for Intelligent Video Analysis and Application of China。
文摘In this study, we address the problems encountered by incremental face clustering. Without the benefit of having observed the entire data distribution, incremental face clustering is more challenging than static dataset clustering. Conventional methods rely on the statistical information of previous clusters to improve the efficiency of incremental clustering;thus, error accumulation may occur. Therefore, this study proposes to predict the summaries of previous data directly from data distribution via supervised learning. Moreover, an efficient framework to cluster previous summaries with new data is explored. Although learning summaries from original data costs more than those from previous clusters, the entire framework consumes just a little bit more time because clustering current data and generating summaries for new data share most of the calculations. Experiments show that the proposed approach significantly outperforms the existing incremental face clustering methods, as evidenced by the improvement of average F-score from 0.644 to 0.762. Compared with state-of-the-art static face clustering methods, our method can yield comparable accuracy while consuming much less time.
