TUM-FAÇADE: REVIEWING AND ENRICHING POINT CLOUD BENCHMARKS FOR FAÇADE SEGMENTATION
نویسندگان
چکیده
Abstract. Point clouds are widely regarded as one of the best dataset types for urban mapping purposes. Hence, point cloud datasets commonly investigated benchmark various interpretation methods. Yet, few researchers have addressed use benchmarks façade segmentation. Robust segmentation is becoming a key factor in applications ranging from simulating autonomous driving functions to preserving cultural heritage. In this work, we present method enriching existing with façade-related classes that been designed facilitate testing. We propose how efficiently extend and comprehensively assess their potential create TUM-FAÇADE dataset, which extends capabilities TUM-MLS-2016. Not only can development point-cloud-based tasks, but our procedure also be applied enrich further datasets.
منابع مشابه
Façade Segmentation in a Multi-view Scenario
We examine a new method of façade segmentation in a multi-view scenario. A set of overlapping, thus redundant street-side images exists and each image shows multiple buildings. A semantic segmentation identifies primary areas in the image such as sky, ground, vegetation, and façade. Subsequently, repeated patterns are detected in image segments previous labeled as “façade areas” and are applied...
متن کاملMorphological segmentation of building façade images
In this paper, we describe an automatic method for segmentation of building façade images. First, individual façades are isolated from general city block images. This step is based on accumulation of directional color gradients, assuming that façade structures are aligned. Then sky region is detected based on segmentation approach and color marker extraction. Finally, the images are split in fl...
متن کاملSymmetry-Based Façade Repair
In this paper we address the problem of removing unwanted image content in a single orthographic façade image. We exploit the regular structure present in building façades and introduce a diffusion process that is guided by the symmetry prevalent in the image. It removes larger unwanted image objects such as traffic lights, street signs, or cables as well as smaller noise, such as reflections i...
متن کاملInteractive Coherence-Based Façade Modeling
We propose a novel interactive framework for modeling building façades from images. Our method is based on the notion of coherence-based editing which allows exploiting partial symmetries across the façade at any level of detail. The proposed workflow mixes manual interaction with automatic splitting and grouping operations based on unsupervised cluster analysis. In contrast to previous work, o...
متن کاملLearning Domain Knowledge for Façade Labelling
This paper presents an approach to address the problem of image façade labelling. In the architectural literature, domain knowledge is usually expressed geometrically in the final design, so façade labelling should on the one hand conform to visual evidence, and on the other hand to the architectural principles – how individual assets (e.g. doors, windows) interact with each other to form a faç...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
سال: 2022
ISSN: ['1682-1777', '1682-1750', '2194-9034']
DOI: https://doi.org/10.5194/isprs-archives-xlvi-2-w1-2022-529-2022