Daniel Cremers Livres

The five-volume set LNCS 9003--9007 constitutes the thoroughly refereed post-conference proceedings of the 12th Asian Conference on Computer Vision, ACCV 2014, held in Singapore, Singapore, in November 2014. The total of 227 contributions presented in these volumes was carefully reviewed and selected from 814 submissions. The papers are organized in topical sections on recognition; 3D vision; low-level vision and features; segmentation; face and gesture, tracking; stereo, physics, video and events; and poster sessions 1-3.

With the rapid advancement of video imaging technology and significant enhancements in CPU speed and memory, video processing and computational video are gaining popularity. This shift mirrors the digital revolution in photography from fifteen years ago, transforming the production of television and movies. The introduction of professional digital movie cameras, digital projectors for theaters, and 3D films is leading to an all-digital production pipeline, creating new opportunities for capturing dynamic scenes, editing footage, and experiencing visual media. This comprehensive survey compiles selected articles from a workshop on Video Processing and Computational Video held at Dagstuhl Castle, Germany, in October 2010. The seminar gathered junior and senior researchers from computer vision, computer graphics, and image communication across academia and industry to tackle challenges in computational video. During the event, 43 researchers worldwide discussed the latest advancements, contemporary challenges, and future directions in imaging, processing, analyzing, modeling, and rendering real-world dynamic scenes. The eight thoroughly revised papers presented were carefully chosen from over 30 lectures, providing a solid overview of computational video and video processing, with a focus on computational photography, video-based rendering, and 3D video.

The content covers a range of advanced topics in discrete optimization and Markov random fields. It explores multi-label moves for MRFs with truncated convex priors, detection and segmentation of independently moving objects from dense scene flow, and efficient global minimization for the multiphase Chan-Vese model of image segmentation. Techniques such as bipartite graph matching on GPU and pose-invariant face matching using MRF energy minimization are discussed. The text also delves into parallel hidden hierarchical fields for multi-scale reconstruction and general search algorithms for energy minimization problems, alongside partial differential equations. Additional topics include complex diffusion on scalar and vector-valued image graphs, coupled super-resolution with non-parametric motion, and optical flow decomposition models. Various approaches to image segmentation and tracking are presented, including hierarchical pairwise segmentation and tracking as segmentation of spatial-temporal volumes. The annotation also covers parameter estimation for marked point processes in object extraction from remote sensing images and robust segmentation techniques. Furthermore, it addresses shape optimization and registration, intrinsic second-order geometric optimization, and image registration under varying illumination. Techniques for inpainting and image denoising, exemplar-based interpolation, and color image restoration using

KlappentextThis book constitutes the thoroughly refereed post-conference proceedings of the International Dagstuhl-Seminar on Statistical and Geometrical Approaches to Visual Motion Analysis, held in Dagstuhl Castle, Germany, in July 2008. The workshop focused on critical aspects of motion analysis, including motion segmentation and the modeling of motion patterns. The aim was to gather researchers who are experts in the different motion tasks and in the different techniques used; also involved were experts in the study of human and primate vision. The 15 revised full papers presented were carefully reviewed and selected from or initiated by the lectures given at the workshop. The papers are organized in topical sections on optical flow and extensions, human motion modeling, biological and statistical approaches, alternative approaches to motion analysis.