Deep neural networks require specific layers to process point clouds, as the scattered and irregular location of 3D points prevents the use of conventional convolutional filters. We introduce the composite layer, a flexible and general alternative to the existing convolutional operators that process 3D point clouds. We design our composite layer to extract and compress the spatial information from the 3D coordinates of points and then combine this with the feature vectors. Compared to mainstream point-convolutional layers such as ConvPoint and KPConv, our composite layer guarantees greater flexibility in network design and provides an additional form of regularization. To demonstrate the generality of our composite layers, we define both a convolutional composite layer and an aggregate version that combines spatial information and features in a nonlinear manner, and we use these layers to implement CompositeNets. Our experiments on synthetic and real-world datasets show that, in both classification, segmentation, and anomaly detection, our CompositeNets outperform ConvPoint, which uses the same sequential architecture, and achieve similar results as KPConv, which has a deeper, residual architecture. Moreover, our CompositeNets achieve state-of-the-art performance in anomaly detection on point clouds. Our code is publicly available at \url{https://github.com/sirolf-otrebla/CompositeNet}.
The Good, the Bad and the Ugly: Evaluating Convolutional Neural Networks for Prohibited Item Detection Using Real and Synthetically Composited X-ray Imagery
FReLU: Flexible Rectified Linear Units for Improving Convolutional Neural Networks
South China University of Technology(华南理工大学)
整流线性单元(ReLU)是深度卷积神经网络广泛使用的激活函数。然而,由于零硬校正,ReLU网络错过了负值的好处。在本文中,我们提出了一个新的激活函数称为\emph{flexible rectified linear unit (FReLU)},以进一步探讨负值的影响。通过将ReLU的校正点重新设计为可学习参数,FReLU扩展了激活输出的状态。当网络成功训练时,FReLU倾向于收敛到负值,这提高了表现力,从而提高了性能。此外,FReLU被设计为简单而有效,没有指数函数,以保持低成本计算。FReLU不依赖于严格的假设条件,可以自适应地应用于各种网络架构。我们在三个标准的图像分类数据集上评估FReLU,包括CIFAR-10,CIFAR-100和ImageNet。实验结果表明,该方法在普通网络和残差网络上都具有较快的收敛速度和较高的性能。
Flexible Coded Distributed Convolution Computing for Enhanced Straggler Resilience and Numerical Stability in Distributed CNNs
College of Computer Science, Chongqing University(重庆大学计算机学院);School of Electrical Engineering, Korea University(韩国大学电气工程学院);School of Electronic Information and Communications, Huazhong University of Science and Technology(华中科技大学电子信息与通信学院);City University of Hong Kong(香港城市大学)
Integrating Flexible Normalization into Mid-Level Representations of Deep Convolutional Neural Networks
University of Miami(迈阿密大学);Computer Science Department(计算机科学系);Coral Gables, FL 33146(科勒尔盖布尔斯,佛罗里达州 33146(注:这是地址,非机构,不应提取,但根据用户提供的文本,此处可能存在格式问题,原文本中“Affiliation: Coral Gables, FL 33146”可能是笔误,正确机构应为“University of Miami”下的部门,此处严格按用户提取范围,仅保留明确机构名))
NullHop: A Flexible Convolutional Neural Network Accelerator Based on Sparse Representations of Feature Maps
Institute of Neuroinformatics, University of Zurich and ETH Zurich(苏黎世大学与苏黎世联邦理工学院神经信息学研究所);Robotic and Tech. of Computers Lab. University of Seville(塞维利亚大学机器人与计算机技术实验室);iniLabs GmbH(ini实验室有限公司)
Cascaded Volumetric Convolutional Network for Kidney Tumor Segmentation from CT volumes
Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所);University of Chinese Academy of Sciences(中国科学院大学);Southeast University(东南大学);Lenovo Research(联想研究院)