Home /

Research

Showing 133 - 138 / 897

RaP-Net: A Region-wise and Point-wise Weighting Network to Extract Robust Keypoints for Indoor Localization


Authors:  DongjiangLi, JinyuMiao, XuesongShi....
Published date-12/01/2020
Tasks:  IndoorLocalization, VisualLocalization

Abstract: Image keypoint extraction is an important step for visual localization. The localization in indoor environment is challenging for that there may be many unreliable features on dynamic or repetitive objects. …

Counting People by Estimating People Flows


Authors:  WeizheLiu, MathieuSalzmann, PascalFua....
Published date-12/01/2020
Tasks:  ActiveLearning, CrowdCounting, OpticalFlowEstimation

Abstract: Modern methods for counting people in crowded scenes rely on deep networks to estimate people densities in individual images. As such, only very few take advantage of temporal consistency in …

Graph Stochastic Neural Networks for Semi-supervised Learning


Authors:  HaiboWang, ChuanZhou, XinChen....
Published date-12/01/2020
Tasks:  NodeClassification, VariationalInference

Abstract: Graph Neural Networks (GNNs) have achieved remarkable performance in the task of the semi-supervised node classification. However, most existing models learn a deterministic classification function, which lack sufficient flexibility to …

Continual Learning of a Mixed Sequence of Similar and Dissimilar Tasks


Authors:  ZixuanKe, BingLiu, XingchangHuang....
Published date-12/01/2020
Tasks:  ContinualLearning, TransferLearning

Abstract: Existing research on continual learning of a sequence of tasks focused on dealing with catastrophic forgetting, where the tasks are assumed to be dissimilar and have little shared knowledge. Some …

Robust compressed sensing using generative models


Authors:  AjilJalal, LiuLiu, AlexandrosG.Dimakis....
Published date-12/01/2020

Abstract: We consider estimating a high dimensional signal in $\R^n$ using a sublinear number of linear measurements. In analogy to classical compressed sensing, here we assume a generative model as a …

Autoencoders that don't overfit towards the Identity


Authors:  HaraldSteck....
Published date-12/01/2020
Tasks:  Denoising

Abstract: Autoencoders (AE) aim to reproduce the output from the input. They may hence tend to overfit towards learning the identity-function between the input and output, i.e., they may predict each …

Filter by

Categories

Tags