在已知三维信息的场景中估计相机位姿,是自主驾驶、增强现实、虚拟现实等领域的重要环节。已有方法从输入图像中直接回归相机的位姿,或者通过回归像素的三维坐标方式计算相机位姿,这些方法存在的问题是与训练场景耦合严重,在新环境中缺...在已知三维信息的场景中估计相机位姿,是自主驾驶、增强现实、虚拟现实等领域的重要环节。已有方法从输入图像中直接回归相机的位姿,或者通过回归像素的三维坐标方式计算相机位姿,这些方法存在的问题是与训练场景耦合严重,在新环境中缺少泛化能力。认为深度学习网络应该专注于学习鲁棒和不变的图像特征,因此介绍了一种基于多尺度图像特征对齐的优化方法,将图像特征相似性作为度量形式,将相机位姿作为优化量,通过从像素到位姿的端到端的训练,来估计相机精确的六自由度(6 degree of freedom,6DOF)位姿。该模型参数和场景分离,对新场景有较强的泛化能力,并且具有较好的定位精度。展开更多
In this study,an underwater image enhancement method based on multi-scale adversarial network was proposed to solve the problem of detail blur and color distortion in underwater images.Firstly,the local features of ea...In this study,an underwater image enhancement method based on multi-scale adversarial network was proposed to solve the problem of detail blur and color distortion in underwater images.Firstly,the local features of each layer were enhanced into the global features by the proposed residual dense block,which ensured that the generated images retain more details.Secondly,a multi-scale structure was adopted to extract multi-scale semantic features of the original images.Finally,the features obtained from the dual channels were fused by an adaptive fusion module to further optimize the features.The discriminant network adopted the structure of the Markov discriminator.In addition,by constructing mean square error,structural similarity,and perceived color loss function,the generated image is consistent with the reference image in structure,color,and content.The experimental results showed that the enhanced underwater image deblurring effect of the proposed algorithm was good and the problem of underwater image color bias was effectively improved.In both subjective and objective evaluation indexes,the experimental results of the proposed algorithm are better than those of the comparison algorithm.展开更多
为了解决教育智能化重构中对于教室人员的高效监督管理和数据分析,结合单阶段目标检测算法的优良特性和卷积神经网络(convolutional neural networks,CNN)良好的特征提取能力,提出了一种基于注意力机制网络和迁移时间序列改进YOLO目标...为了解决教育智能化重构中对于教室人员的高效监督管理和数据分析,结合单阶段目标检测算法的优良特性和卷积神经网络(convolutional neural networks,CNN)良好的特征提取能力,提出了一种基于注意力机制网络和迁移时间序列改进YOLO目标检测算法的教室人员目标检测算法.首先,对源视频流进行逐帧抽取和非畸变的图像放缩,通过生成对抗网络(generative adversarial network,GAN)进行图像超分辨处理;其次,对每帧图像进行多尺度采样和初步目标检测;然后,根据不同尺度得到的候选结果进行非极大抑制(non maximum suppression,NMS)以去除置信度较低的个体;之后,对候选结果进行融合,再使用交并比(intersection over union,IoU)进行重叠度计算以更新数据、去除重合或过于紧密的定位位置,然后将当前帧的检测结果与先前时间区间中的检测结果作为时间序列进行统计学数据迁移融合(time series migration,TSM)获得最后的检测结果.实验结果表明,本文方法不仅有效地提升了教室人员目标检测的准确率,并且可以进行实时检测.展开更多
文摘在已知三维信息的场景中估计相机位姿,是自主驾驶、增强现实、虚拟现实等领域的重要环节。已有方法从输入图像中直接回归相机的位姿,或者通过回归像素的三维坐标方式计算相机位姿,这些方法存在的问题是与训练场景耦合严重,在新环境中缺少泛化能力。认为深度学习网络应该专注于学习鲁棒和不变的图像特征,因此介绍了一种基于多尺度图像特征对齐的优化方法,将图像特征相似性作为度量形式,将相机位姿作为优化量,通过从像素到位姿的端到端的训练,来估计相机精确的六自由度(6 degree of freedom,6DOF)位姿。该模型参数和场景分离,对新场景有较强的泛化能力,并且具有较好的定位精度。
文摘In this study,an underwater image enhancement method based on multi-scale adversarial network was proposed to solve the problem of detail blur and color distortion in underwater images.Firstly,the local features of each layer were enhanced into the global features by the proposed residual dense block,which ensured that the generated images retain more details.Secondly,a multi-scale structure was adopted to extract multi-scale semantic features of the original images.Finally,the features obtained from the dual channels were fused by an adaptive fusion module to further optimize the features.The discriminant network adopted the structure of the Markov discriminator.In addition,by constructing mean square error,structural similarity,and perceived color loss function,the generated image is consistent with the reference image in structure,color,and content.The experimental results showed that the enhanced underwater image deblurring effect of the proposed algorithm was good and the problem of underwater image color bias was effectively improved.In both subjective and objective evaluation indexes,the experimental results of the proposed algorithm are better than those of the comparison algorithm.
文摘为了解决教育智能化重构中对于教室人员的高效监督管理和数据分析,结合单阶段目标检测算法的优良特性和卷积神经网络(convolutional neural networks,CNN)良好的特征提取能力,提出了一种基于注意力机制网络和迁移时间序列改进YOLO目标检测算法的教室人员目标检测算法.首先,对源视频流进行逐帧抽取和非畸变的图像放缩,通过生成对抗网络(generative adversarial network,GAN)进行图像超分辨处理;其次,对每帧图像进行多尺度采样和初步目标检测;然后,根据不同尺度得到的候选结果进行非极大抑制(non maximum suppression,NMS)以去除置信度较低的个体;之后,对候选结果进行融合,再使用交并比(intersection over union,IoU)进行重叠度计算以更新数据、去除重合或过于紧密的定位位置,然后将当前帧的检测结果与先前时间区间中的检测结果作为时间序列进行统计学数据迁移融合(time series migration,TSM)获得最后的检测结果.实验结果表明,本文方法不仅有效地提升了教室人员目标检测的准确率,并且可以进行实时检测.
