CROwdsourced and COmputational rephotoGRAPHY - accurately mapping images of the past on the buildings and landscapes of today

CroCoGraphy 2018


Computer Graphics Computer Vision & Pattern Recognition



SCOPE:
Crowdsourced and computational rephotography is a multidisciplinary subject, dealing with a broad range of computer vision, metadata, semantics and VGI techniques to estimate the viewpoint of old photographs in today’s images of the world. Aligning a historical photograph in a modern rephotograph (or vice versa) can serve as a remarkable visualization of the passage of time and can be used across a plethora of application domains (e.g., tourism, media and GLAM sector). The workshop welcomes contributions to theories, design and evaluation of volunteered and automatic rephotography (or a combination of both) and will provide an interactive platform with presentation and discussion opportunities for novel research, work in progress, and more general 'open problems' in rephotography and related research domains.
Topics:
Volunteered geographic information (VGI) and crowdsourcing rephotography Computer vision techniques for computational rephotography Semantic scene understanding Clustering of visually similar landscapes/scenes Architectural style/building type recognition Image geolocalization Geographic metadata enrichment and metadata-driven image analysis Computational efficient rephotography visualization techniques Visual Place Recognition in Changing Environments / Change invariant scene matching (Mobile) Imaging for Cultural Heritage Learning invariant places / image features