Comparison of Training Strategies for Convnets on Multiple Similar Datasets for Facade Segmentation

Schmitz, Matthias; Huang, Hai; Mayer, Helmut

doi:10.5194/isprs-archives-xlii-2-w13-111-2019

Cited by 3 publications

(2 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Building façade segmentation dates back a few decades. The current cutting edge of both building façade and the broader urban scene segmentation studies relies mostly on deploying and expanding methods using convolutional neural networks of various architectures (Badrinarayanan et al., 2016; Femiani et al., 2018; Fu et al., 2019; Zhao et al., 2017; Zolanvari et al., 2018; Schmitz et al., 2019). The community has also developed a series of public and often‐used test datasets including ECP (Teboul et al., 2010) and Graz (Riemenschneider et al., 2012) for façade segmentation and Cityscapes (Cordts et al., 2016), Mapillary Vistas (Neuhold et al., 2017), and ApolloScape (X. Huang et al., 2018) for urban scene segmentation.…”

Section: Building Stock Characterization: Relevance Challenges and Op...mentioning

confidence: 99%

A scalable data collection, characterization, and accounting framework for urban material stocks

Arbabi

Lanau

Meyers

et al. 2021

J of Industrial Ecology

View full text Add to dashboard Cite

Building stocks represent an extensive reservoir of secondary resources. However, common bottom‐up characterization of these, often based on archetypal classification of buildings and their corresponding material intensity, are still not suitable to adequately inform circular economic strategies. Indeed, these approaches typically result in a loss of building‐specific details, and a building stock characterization in terms of material mass, for example, glass, rather than component, for example, window. To deliver this higher resolution of details, a scalable approach to urban stock characterization, that enables a bottom‐up estimation of building stocks at the building component level, is needed. In this paper, we present a framework to automate the characterization of urban stock. By using and combining a mobile‐sensing approach with computer vision, urban stocks can be captured as 3D surface maps allowing the identification and semantic classification of stock objects, components, and materials. We demonstrate the potential of this framework through a case study of a neighborhood in Sheffield, UK, by using a prototype workflow comprising a custom‐made mobile‐sensing platform and an existing suite of neural network models to calculate an estimate count of buildings external doors and windows. The prototype implementation of the framework achieves comparable total and building‐level component counts with those achieved through manual human counts. Such automated estimation of components enables an understanding of opportunities across the circular economic hierarchies and informs stakeholders across the supply chain to better prepare for the implementation of circular strategies including building refurbishments.

show abstract

Section: Building Stock Characterization: Relevance Challenges and Op...mentioning

confidence: 99%

A scalable data collection, characterization, and accounting framework for urban material stocks

Arbabi

Lanau

Meyers

et al. 2021

J of Industrial Ecology

View full text Add to dashboard Cite

show abstract

“…Semantic segmentation is a Computer Vision technique (Förstner and Wrobel, 2016) that aims to the recognition and the comprehension of the content of an image at the pixel level. This approach is widely used in remote sensing applications, especially in the analysis of urban scenarios (Ajmar et al, 2019;Huang et al, 2019, Schmitz et al, 2019, Zhou et al, 2019 or in the delineation of forest trees (Chen et al, 2021;Sothe et al, 2020;Kempf et al, 2019). The segmentation approach could be based on imagery (Marmanis et al, 2018) or three-dimensional models (Ao et al, 2019), as well as on the combination of both 2D and 3D information (Ding et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Mago Approach for Semantic Segmentation: The Case Study of Uavid Benchmark Dataset

Gagliolo

Sguerso

2021

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. The present work is focused on a semantic segmentation strategy implemented in the workflow of the tool MAGO (standing for “Adaptive Mesh for Orthophoto Generation”), considering the contribution of the 3D geometry and the colour information, both deriving from the point cloud of the scene. Moreover, the 2D source imagery, previously used to obtain the photogrammetric point cloud, is employed even to enhance the procedure with the recognition of moving objects, comparing the evolution of epochs.The analysed context is an urban scene, deriving from the UAVid dataset proposed for the ISPRS benchmark. In particular, the so-called “seq18”, a set of high-resolution oblique images taken by UAV (Unmanned Aerial Vehicle), has been used to test the semantic segmentation. The workflow includes the production of two Digital Surface Models (DSMs), containing the geometric and radiometric information, respectively, and their processing by means of the Harris corner detector, allowing the understanding of the image variability. Then, starting from the source geometry and colour information and combining them with their variability mapping, a preliminary classification is performed. Further criteria allow the segmentation of the humans and cars present in the scene. In particular, static objects are identified according to the content of the neighbour pixels in a certain kernel, while the evolution in time of moving elements is recognized by means of the comparison of the projected images belonging to the different epochs. The presented preliminary achievements show some criticalities that require further attention and improvement. In particular, the strategy could be enriched getting more information from the source 2D images, which at the moment are directly used only for the comparison of consecutive epochs.

show abstract

Lod3 Building Reconstruction From Multi-Source Images

Huang

Michelini

Schmitz

et al. 2020

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. We propose a pipeline for the detection as well as modeling of individual buildings based on multi-source images. It allows to consistently reconstruct whole buildings at Level of Detail 3 (LoD3): the roof from airborne images and the facades including elements such as windows and doors mainly from terrestrial images. We employ a parametrized top-down model – the “shell model” – with the roof as well as the facades semantically and geometrically integrated. This generative model fosters stability for building detection by enabling the use of multi-source data and offers flexibility in modeling by means of a fully CAD-compatible integration of building components. Experiments performed on imagery from different terrestrial and airborne (Unmanned Aerial Vehicle – UAV) cameras demonstrate the potential of the approach.

show abstract

Comparison of Training Strategies for Convnets on Multiple Similar Datasets for Facade Segmentation

Cited by 3 publications

References 18 publications

A scalable data collection, characterization, and accounting framework for urban material stocks

A scalable data collection, characterization, and accounting framework for urban material stocks

Mago Approach for Semantic Segmentation: The Case Study of Uavid Benchmark Dataset

Lod3 Building Reconstruction From Multi-Source Images

Contact Info

Product

Resources

About