TGV-Fusion

Pock, Thomas; Zebedin, Lukas; Bischof, Horst

doi:10.1007/978-3-642-19391-0_18

Cited by 36 publications

(46 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to the TV-functional [2] we recall that constant functions are in the kernel of TV, while polynomials of degree strictly less than k constitute the kernel of TGV k ⃗ α ; see [1]. Since its introduction, TGV k ⃗ α functionals have proved to be effective in diverse mathematical imaging problems, including denoising [1], reconstruction of magnetic resonance images from highly undersampled data [3], deconvolution [4] and fusion of stereographical data [5]. This success motivates, besides inherent mathematical interest, to further investigate analytical properties of these functionals.…”

Section: The L 1 -Tgv 2 Functionalmentioning

confidence: 99%

Properties ofL1-TGV2: The one-dimensional case

Bredies

Kunisch

Valkonen

2013

Journal of Mathematical Analysis and Applications

View full text Add to dashboard Cite

show abstract

Section: The L 1 -Tgv 2 Functionalmentioning

confidence: 99%

Properties ofL1-TGV2: The one-dimensional case

Bredies

Kunisch

Valkonen

2013

Journal of Mathematical Analysis and Applications

View full text Add to dashboard Cite

show abstract

“…The main requirement for such a smoothing algorithm is that it minimizes noise but preserves sharp edges and small structures. For this reason and to accelerate the plane detection we smooth the data using the "total generalized variation" TGV [13]. This method has the advantage that it can be efficiently implemented on the GPU and is therefore very fast and robust.…”

Section: Dsm Smoothingmentioning

confidence: 99%

“…First is a smoothing of the DSM with a total generalized variation method (TGV) [13]. Then all roof planes get extracted using an approach first introduced by [14] and employing so-called random sampling and conceptual representation.…”

Section: Approachmentioning

confidence: 99%

Planar roof surface segmentation using 3D vision

Meixner

Leberl

Brédif

2011

Proceedings of the 19th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems

View full text Add to dashboard Cite

Internet search has initially been a strong driving force for the rapid emergence of 3D building models of large urban areas. Additionally, many commercial and governmental initiatives have been started to develop urban 3D geographic information systems in a transition from the classical 2D-to the novel 3D-GIS. The modeling of building roofs is thus a relevant research topic. The focus has been on the use of aerial LiDAR point clouds (Light Detection And Ranging). However, recent progress in digital aerial cameras has rendered possible the acquisition of very dense point clouds from high overlap digital aerial imagery, and to use these point clouds jointly with the image information to generate 3D building models. This paper presents a multi-step processing framework and work flow for the automatic segmentation of building roofs in densely built-up areas from high-resolution vertical aerial images. Details extruding from, or intruding into, a roof are being excluded so that each roof is being modeled by means of its planar segments and can then be classified as a specific roof type from a set of standard roof shapes. Our experimental work employs a test area in Graz (Austria) with 186 buildings.

show abstract

“…A more advanced technique was proposed by Papasaika [25], in which sparse representation supported by weights served for fusion of DEMs from various data sources. Pock et al [26] proposed Total Generalized Variational (TGV) methods for fusion of airborne optical-stereoscopic DEMs, while a weighted version of total variational (TV) method and TGV were examined by Kuschk et al [27] on different space borne optical DEMs. Fuss et al utilized the modified K-means clustering algorithm to fuse multiple overlapping radargrammetric Envisat-2 DEMs [28].…”

Section: Introductionmentioning

confidence: 99%

Fusion of TanDEM-X and Cartosat-1 elevation data supported by neural network-predicted weight maps

Bagheri

Schmitt

Zhu

2018

ISPRS Journal of Photogrammetry and Remote Sensing

View full text Add to dashboard Cite

This is the pre-acceptance version, to read the final version, please go to ISPRS Journal of Photogrammetry and Remote Sensing on ScienceDirect. Recently, the bistatic SAR interferometry mission TanDEM-X provided a global terrain map with unprecedented accuracy. However, visual inspection and empirical assessment of TanDEM-X elevation data against high-resolution ground truth illustrates that the quality of the DEM decreases in urban areas because of SAR-inherent imaging properties. One possible solution for an enhancement of the TanDEM-X DEM quality is to fuse it with other elevation data derived from highresolution optical stereoscopic imagery, such as that provided by the Cartosat-1 mission. This is usually done by Weighted Averaging (WA) of previously aligned DEM cells. The main contribution of this paper is to develop a method to efficiently predict weight maps in order to achieve optimized fusion results. The prediction is modeled using a fully connected Artificial Neural Network (ANN). The idea of this ANN is to extract suitable features from DEMs that relate to height residuals in training areas and then to automatically learn the pattern of the relationship between height errors and features. The results show the DEM fusion based on the ANN-predicted weights improves the qualities of the study DEMs. Apart from increasing the absolute accuracy of Cartosat-1 DEM by DEM fusion, the relative accuracy (respective to reference LiDAR data) of DEMs is improved by up to 50% in urban areas and 22% in non-urban areas while the improvement by the HEM-based method does not exceed 20% and 10% in urban and non-urban areas respectively.

show abstract

TGV-Fusion

Cited by 36 publications

References 10 publications

Properties ofL1-TGV2: The one-dimensional case

Properties ofL1-TGV2: The one-dimensional case

Planar roof surface segmentation using 3D vision

Fusion of TanDEM-X and Cartosat-1 elevation data supported by neural network-predicted weight maps

Contact Info

Product

Resources

About