State-of-the-arts studies on sentiment classification are typically domain-dependent and domain-restricted. In this paper, we aim to reduce domain dependency and improve overall performance simultaneously by proposing...State-of-the-arts studies on sentiment classification are typically domain-dependent and domain-restricted. In this paper, we aim to reduce domain dependency and improve overall performance simultaneously by proposing an efficient multi-domain sentiment classification algorithm. Our method employs the approach of multiple classifier combination. In this approach, we first train single domain classifiers separately with domain specific data, and then combine the classifiers for the final decision. Our experiments show that this approach performs much better than both single domain classification approach （using the training data individually） and mixed domain classification approach （simply combining all the training data）. In particular, classifier combination with weighted sum rule obtains an average error reduction of 27.6% over single domain classification.展开更多
为了适应视觉跟踪过程中目标外观变化,提高视觉跟踪算法的鲁棒性,本文基于卷积神经网络(Convolutional Neural Network,CNN)并结合多域学习法与多模板管理,提出一种通过树形结构管理多模板的多域卷积神经网络(Multi-Domain CNNs with Mu...为了适应视觉跟踪过程中目标外观变化,提高视觉跟踪算法的鲁棒性,本文基于卷积神经网络(Convolutional Neural Network,CNN)并结合多域学习法与多模板管理,提出一种通过树形结构管理多模板的多域卷积神经网络(Multi-Domain CNNs with Multiple Models in a tree structure)视觉跟踪算法。首先使用大量已标记目标位置的视频数据预训练多域结构的CNN,使CNN卷积层可从图像中提取出适用于跟踪任务的特征。然后在跟踪时中对CNN全连接层进行微调以适应跟踪目标,并使用树形结构管理存储不同时间段的目标模板得到模板树。使用模板树综合评价待检测帧,估计目标位置。最后按照一定规则将新模板添加进模板树,完成模板的更新。实验表明,该算法对跟踪过程中目标外观的变化有着良好的适应性,同时多模板可抑制CNN在跟踪时产生的模板漂移问题。展开更多
基金Supported by the National Natural Science Foundation of China under Grant No.61003155Start-Up Grant for Newly Appointed Professors under Grant No.1-BBZM in The Hong Kong Polytechnic University
文摘State-of-the-arts studies on sentiment classification are typically domain-dependent and domain-restricted. In this paper, we aim to reduce domain dependency and improve overall performance simultaneously by proposing an efficient multi-domain sentiment classification algorithm. Our method employs the approach of multiple classifier combination. In this approach, we first train single domain classifiers separately with domain specific data, and then combine the classifiers for the final decision. Our experiments show that this approach performs much better than both single domain classification approach （using the training data individually） and mixed domain classification approach （simply combining all the training data）. In particular, classifier combination with weighted sum rule obtains an average error reduction of 27.6% over single domain classification.
文摘为了适应视觉跟踪过程中目标外观变化,提高视觉跟踪算法的鲁棒性,本文基于卷积神经网络(Convolutional Neural Network,CNN)并结合多域学习法与多模板管理,提出一种通过树形结构管理多模板的多域卷积神经网络(Multi-Domain CNNs with Multiple Models in a tree structure)视觉跟踪算法。首先使用大量已标记目标位置的视频数据预训练多域结构的CNN,使CNN卷积层可从图像中提取出适用于跟踪任务的特征。然后在跟踪时中对CNN全连接层进行微调以适应跟踪目标,并使用树形结构管理存储不同时间段的目标模板得到模板树。使用模板树综合评价待检测帧,估计目标位置。最后按照一定规则将新模板添加进模板树,完成模板的更新。实验表明,该算法对跟踪过程中目标外观的变化有着良好的适应性,同时多模板可抑制CNN在跟踪时产生的模板漂移问题。
