Indoor scene texturing based on single mobile phone images and 3D model fusion

Realistic texture mapping and coherent up-to-date rendering is one of the most important issues in indoor 3-D modelling.However,existing texturing approaches are usually performed manually during the modelling process,and cannot accommodate changes in indoor environments occurring after the model was created,resulting in outdated and misleading texture rendering.In this study,a structured learning-based texture mapping method is proposed for automatic mapping a single still photo from a mobile phone onto an alreadyconstructed indoor 3-D model.The up-to-date texture is captured using a smart phone,and the indoor structural layout is extracted by incorporating per-pixel segmentation in the FCN algorithm and the line constraints into a structured learning algorithm.This enables real-time texture mapping according to parts of the model,based on the structural layout.Furthermore,the rough camera pose is estimated by pedestrian dead reckoning(PDR)and map information to determine where to map the texture.The experimental results presented in this paper demonstrate that our approach can achieve accurate fusion of 3-D triangular meshes with 2-D single images,achieving low-cost and automatic indoor texture updating.Based on this fusion approach,users can have a better experience in virtual indoor3-D applications.

Imaging volumes during COVID-19:A Victorian health service experience

BACKGROUND The World Health Organisation declared the coronavirus disease 2019(COVID-19)a pandemic on March 11,2020.While globally,the relative caseload has been high,Australia’s has been relatively low.During the pandemic,radiology services have seen significant changes in workflow across modalities and a reduction in imaging volumes.AIM To investigate differences in modality imaging volumes during the COVID-19 pandemic across a large Victorian public health network.METHODS A retrospective analysis from January 2019 to December 2020 compared imaging volumes across two periods corresponding to the pandemic’s first and second waves.Weekly volumes across patient class,modality and mobile imaging were summed for periods:wave 1(weeks 11 to 16 for 2019;weeks 63 to 68 for 2020)and wave 2(weeks 28 to 43 for 2019;weeks 80 to 95 for 2020).Microsoft Power Business Intelligence linked to the radiology information system was used to mine all completed examinations.RESULTS Summed weekly data during the pandemic’s first wave showed the greatest decrease of 29.8%in adult outpatient imaging volumes and 46.3%in paediatric emergency department imaging volumes.Adult nuclear medicine demonstrated the greatest decrease of 37.1%for the same period.Paediatric nuclear medicine showed the greatest decrease of 47.8%,with angiography increasing by 50%.The pandemic’s second wave demonstrated the greatest decrease of 23.5%in adult outpatient imaging volumes,with an increase of 18.2%in inpatient imaging volumes.The greatest decrease was 28.5%in paediatric emergency department imaging volumes.Nuclear medicine showed the greatest decrease of 37.1%for the same period.Paediatric nuclear medicine showed the greatest decrease of 36.7%.Mobile imaging utilisation increased between 57.8%and 135.1%during the first and second waves.A strong correlation was observed between mobile and nonmobile imaging in the emergency setting(Spearman’s correlation coefficient=-0.743,P=0.000).No correlation was observed in the inpatient setting(Spearman’s correlation coefficient=-0.059,P=0.554).CONCLUSION Nuclear medicine was most impacted,while computed tomography and angiography were the least affected by the pandemic.The impact was less during the pandemic’s second wave.Mobile imaging shows continuous growth during both waves.

Visual Person Identification Using a Distance-dependent Appearance Model for a Person Following Robot

This paper describes a person identifcation method for a mobile robot which performs specifc person following under dynamic complicated environments like a school canteen where many persons exist.We propose a distance-dependent appearance model which is based on scale-invariant feature transform（SIFT） feature.SIFT is a powerful image feature that is invariant to scale and rotation in the image plane and also robust to changes of lighting condition.However,the feature is weak against afne transformations and the identifcation power will thus be degraded when the pose of a person changes largely.We therefore use a set of images taken from various directions to cope with pose changes.Moreover,the number of SIFT feature matches between the model and an input image will decrease as the person becomes farther away from the camera.Therefore,we also use a distance-dependent threshold.The person following experiment was conducted using an actual mobile robot,and the quality assessment of person identifcation was performed.

Investigation of multilayer domains in large-scale CVD monolayer graphene by optical imaging

CVD graphene is a promising candidate for optoelectronic applications due to its high quality and high yield.However,multi-layer domains could inevitably form at the nucleation centers during the growth.Here,we propose an optical imaging technique to precisely identify the multilayer domains and also the ratio of their coverage in large-scale CVD monolayer graphene.We have also shown that the stacking disorder in twisted bilayer graphene as well as the impurities on the graphene surface could be distinguished by optical imaging.Finally,we investigated the effects of bilayer domains on the optical and electrical properties of CVD graphene,and found that the carrier mobility of CVD graphene is seriously limited by scattering from bilayer domains.Our results could be useful for guiding future optoelectronic applications of large-scale CVD graphene.