Konrad Wenzel

4D Reconstruction of Tangible Cultural Heritage Objects from Web-Retrieved Images

The number of digital images that are available online today has reached unprecedented levels. Recent statistics showed that by the end of 2013 there were over 250 billion photographs stored in just one of the major social media sites, with a daily average upload of 300 million photos. These photos, apart from documenting personal lives, often relate to experiences in well-known places of cultural interest, throughout several periods of time. Thus from the viewpoint of Cultural Heritage professionals, they constitute valuable and freely available digital cultural content. Advances in the fields of Photogrammetry and Computer Vision have led to significant breakthroughs such as the Structure from Motion algorithm which creates 3D models of objects using their 2D photographs. The existence of powerful and affordable computational machinery enables the reconstruction not only of single structures such as artefacts, but also of entire cities. This paper presents an overview of our methodology for producing co...

Fast and Robust Generation of Semantic Urban Terrain Models from UAV Video Streams

We present an algorithm for extracting Level of Detail 2 (LOD2) building models from video streams captured by Unmaned Aerial Vehicles (UAVs). Typically, such imagery is of limited radiometric quality but the surface is captured with large redundancy. The first contribution of this paper is a novel algorithm exploiting this redundancy for precise depth computation. This is realized by fusing consistent depth estimations across single stereo models and generating a 2.5D elevation map from the resulting point clouds. Disparity maps are derived by a coarse-to-fine Semi-Global-Matching (SGM) method performing well on noisy imagery. The second contribution concerns a challenging step of the context-based urban terrain modeling: Dominant planes extraction for building reconstruction. Because of noisy data and complicated roof structures, both dominant plane parameters and initial values for support sets of planes are obtained by the J-Linkage algorithm. An improved point-to-plane labeling is presented to encourage the assignment of proximate points to the same plane. This is accomplished by non-local, Markov Random Field (MRF) - based optimization and segmentation of color information. The potential and the limitations of the proposed methods are shown using an UAV video sequence of limited radiometric quality.

Photogrammetric point cloud collection with multi-camera systems

We present an efficient method for the recording of 3D point clouds using a compact handheld camera rig and an automated software pipeline for an accurate surface reconstruction. Multiple industrial cameras are mounted on a rectangular shaped frame with a size of 15cm by 15cm in order to collect images from multiple views at once. By using the presented software pipeline, one dense 3D point cloud can be computed efficiently for each shot. The system is particularly designed for close range cultural heritage applications, where the requirements regarding accuracy, density but also acquisition efficiency are high. For each shot up to 3.5 Mio. 3D points can be derived. An area of 60cm by 50cm is covered at a distance of 70cm. Depending on distance and surface texture the points reach a precision of up to 0.2mm. Within this paper, we will present the system design, the data acquisition process, the automatic orientation/registration approach and the dense surface reconstruction method. Finally, we will demonstrate results for an example covering a large scale cultural heritage project, where 2 billion 3D points were acquired efficiently with sub-mm accuracy.

Complementing TLS Point Clouds by Dense Image Matching

The complementation of terrestrial laser scanning (TLS) by photogrammetric acquisition techniques enables the exploitation of two different measurement principles. TLS with distance measurements can be used for acquiring large-scale point clouds at medium range distances, while image-based surface reconstruction methods enable flexible acquisition with high precision at short distances. Within this paper, an automatic procedure for the combination of photogrammetry and laser scanning is presented. Synthetic images derived from the laser scanning point clouds are used together with camera images in a common Structure-from- Motion (SfM) process. This enables an automatic registration of multiple laser scans. Furthermore, surface information can also be derived from the imagery using dense image matching methods. While the laser scanning data provides the missing scale information, the imagery can complement the dataset to fill gaps, occlusions or to resolve small details. Consequently, large datasets can be...

4D reconstruction of the past: the image retrieval and 3D model construction pipeline

One of the main characteristics of the Internet era we are living in, is the free and online availability of a huge amount of data. This data is of varied reliability and accuracy and exists in various forms and formats. Often, it is cross-referenced and linked to other data, forming a nexus of text, images, animation and audio enabled by hypertext and, recently, by the Web3.0 standard. Our main goal is to enable historians, architects, archaeolo- gists, urban planners and affiliated professionals to reconstruct views of historical monuments from thousands of images floating around the web. This paper aims to provide an update of our progress in designing and imple- menting a pipeline for searching, filtering and retrieving photographs from Open Access Image Repositories and social media sites and using these images to build accurate 3D models of archaeological monuments as well as enriching multimedia of cultural / archaeological interest with metadata and harvesting the end products to EU- ROPEANA. We provide details of how our implemented software searches and retrieves images of archaeological sites from Flickr and Picasa repositories as well as strategies on how to filter the results, on two levels; a) based on their built-in metadata including geo-location information and b) based on image processing and clustering techniques. We also describe our implementation of a Structure from Motion pipeline designed for producing 3D models using the large collection of 2D input images (>1000) retrieved from Internet Repositories.

Automated and Accurate Orientation of Large Unordered Image Datasets for Close-Range Cultural Heritage Data Recording Automatische und hochgenaue Orientierung von großen, ungeordneten Bildverbänden bei der photogrammetrischen Aufnahme von Weltkulturdenkmälern

Summary: Reconstruction of image orientationsand geometry from images is one of the basic tasksin photogrammetry and computer vision. A fullyautomated solution of this task in terrestrial appli-cations is still pending in case of large unorderedimage datasets especially for close-range and/orlow-cost applications. Current solutions requirehigh computationaleffortsforimagenetworkswithhigh complexity and diversity regarding acquisi-tion geometry. Unlike the methods suitable forlandmark reconstruction from large-scale Internetimage collections we focus on datasets where onecannot reduce the number of images without losinggeometric information of the dataset. Within thepaper, an automated pipeline for the reconstructionof reliable and precise camera orientation from un-ordered image datasets is presented. Results for aclose-rangeculturalheritageapplication, theexam-ple ofthe Amsterdam project, are shown to demon-strate the performance ofthe presented pipeline forapplications with low cost and high accuracy re-quirements.