ID 原文 译文
38926 随后构建融合对称特性的混合标签传递半监督直觉模糊聚类目标函数，通过聚类获得最终的分割结果。 Fourth, it constructed a hybrid label propagation semi-supervised intuitionistic fuzzy clustering objective function incorporating symmetric property. Finally, the final segmentation result was gained by the proposed algorithm.
38927 两个彩色图像库上的实验结果表明，该算法能够将目标从复杂背景中完整的分割出来，分割性能优于对比算法。 Experimental results on two color image libraries demonstrated that the proposed algorithm could segment the target from the complex background completely, and the segmentation performance was superior to the comparison algorithms.
38928 在视觉跟踪应用中，目标外观通常由包含目标的矩形区域来建模，这种矩形化边框的描述方式不可避免地引入了背景干扰，并随着场景变化导致跟踪关注点的模糊及歧义，进而产生跟踪漂移。 In visual object tracking, the appearance of the target is usually modeled by a bounding-box containing the target, which inevitably introduces background interference. As the scene changes, the concerns become blurred and ambiguous, and then produces tracking drift.
38929 针对以上问题，提出了一种基于多注意力图的孪生网络视觉目标跟踪算法。 Considering the above problems, a Siamese network with multi-attention map for visual object tracking is proposed.
38930 首先，建立了一种关注于前景目标区域特征表达的孪生网络。 Firstly, a Siamese network focusing on foreground feature representation of target is established.
38931 该网络通过构建梯度注意力图损失函数项来引导网络训练，提升网络区分目标和干扰背景的能力。 The gradient attention loss function is constructed to guide network training and improve the ability of distinguishing target and interference background.
38932 此外，嵌入通道注意力和空间注意力进一步强化目标的特征表达，自动发掘有区分的特征表示。 In addition, embedding channel attention and spatial attention further strengthens the feature expression of the target, and automatically discovers the distinguished feature expression.
38933 在多个公共数据集上的实验验证了提出算法的有效性，以及算法可完成实时的视觉目标跟踪。 Extensive experiments on benchmark datasets demonstrate that the proposed tracker performs favorably, and its ability to achieve real-time visual object tracking.
38934 为了提高视频编码的容错性能，保证视频经不可靠信道传输后的重建质量。本文提出了一种面向高效视频编码标准(High Efficiency Video Coding，HEVC)的基于参数重用的多描述视频编码方法。 In order to improve the error resilient performance of video coding, and guarantee the reconstructed quality of video transmitted through unreliable channel, a multiple description video coding method based on parameter reuse for High Efficiency Video Coding(HEVC) is proposed in this paper.
38935 原始视频进行空间梅花下采样，生成四个行列交错的子序列，其中两个子序列采用标准编码器进行编码，并在编码过程中提取视频中每个编码单元(Coding Unit，CU)的深度信息、预测单元(Predicting Unit，PU)的分割方式以及帧内预测模式。而其余两个子序列利用已编码的视频序列信息，进行简化的编码过程。 Polyphase down-sampling is used to generate four sub-sequences with interleaved rows and columns, two of which are encoded by a standard HEVC encoder.Meanwhile, the depth information of each coding unit(CU), the partition of predicting unit(PU) and intra prediction mode are extracted during the encoding process, which are reused by the other two subsequences to simplify the coding process.