CumInCAD is a Cumulative Index about publications in Computer Aided Architectural Design
supported by the sibling associations ACADIA, CAADRIA, eCAADe, SIGraDi, ASCAAD and CAAD futures

PDF papers
id ecaadesigradi2019_339
authors Kinugawa, Hina and Takizawa, Atsushi
year 2019
title Deep Learning Model for Predicting Preference of Space by Estimating the Depth Information of Space using Omnidirectional Images
source Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 2, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 61-68
summary In this study, we developed a method for generating omnidirectional depth images from corresponding omnidirectional RGB images of streetscapes by learning each pair of omnidirectional RGB and depth images created by computer graphics using pix2pix. Then, the models trained with different series of images shot under different site and weather conditions were applied to Google street view images to generate depth images. The validity of the generated depth images was then evaluated visually. In addition, we conducted experiments to evaluate Google street view images using multiple participants. We constructed a model that estimates the evaluation value of these images with and without the depth images using the learning-to-rank method with deep convolutional neural network. The results demonstrate the extent to which the generalization performance of the streetscape evaluation model changes depending on the presence or absence of depth images.
keywords Omnidirectional image; depth image; Unity; Google street view; pix2pix; RankNet
series eCAADeSIGraDi
full text file.pdf (10,552,134 bytes)
references Content-type: text/plain
Details Citation Select
100%; open Benedikt, M (1979) Find in CUMINCAD To take hold of space: isovists and isovist fields , Environment and Planning B, 6, pp. 47-65

100%; open Burges, C, et al. (2005) Find in CUMINCAD Learning to rank using gradient descent , ICML, pp. 89-96

100%; open Chen, LC, et al. (2018) Find in CUMINCAD Encoder-decoder with atrous separable convolution for semantic image segmentation , ECCV2018

100%; open He, K, et al. (2016) Find in CUMINCAD Deep residual learning for image recognition , CVPR2016, pp. 770-778

100%; open Hillier, B and Hanson, J (1989) Find in CUMINCAD The Social Logic of Space , Cambridge University Press

100%; open Isola, P, et al. (2017) Find in CUMINCAD Image-to-image translation with conditional adversarial networks , CVPR2017, pp. 5967-5976

100%; open Law, S, et al. (2017) Find in CUMINCAD An application of convolutional neural network in street image classification , GeoAI, pp. 5-9

100%; open Liu, L, et al. (2017) Find in CUMINCAD A machine learning-based method for the large-scale evaluation of the urban environment , Computers, Environment and Urban Systems, 65, pp. 113-125

100%; open Ros, G, et al. (2016) Find in CUMINCAD The SYNTHIA dataset: a large collection of synthetic images for semantic segmentation of urban scenes , CVPR2016, pp. 3234-3243

100%; open Takizawa, A and Furuta, A (2017) Find in CUMINCAD 3D spatial analysis method with first-person viewpoint by deep convolutional neural network with omnidirectional RGB and depth images , Proceedings of eCAADe 2017, pp. 693-702

last changed 2019/08/26 20:26
pick and add to favorite papersHOMELOGIN (you are user _anon_54384 from group guest) CUMINCAD Papers Powered by SciX Open Publishing Services 1.002