Computer Vision - ACCV 2012 [electronic resource] : 11th Asian Conference on Computer Vision, Daejeon, Korea, November 5-9, 2012, Revised Selected Papers, Part IV / edited by Kyoung Mu Lee, Yasuyuki Matsushita, James M. Rehg, Zhanyi Hu.

Oral Session 8: Shape Reconstruction and Optimization -- Self-calibration and Motion Recovery from Silhouettes with Two Mirrors -- Stereo Reconstruction and Contrast Restoration in Daytime Fog -- Large-Scale Bundle Adjustment by Parameter Vector Partition -- Learning Feature Subspaces for Appearance-Based Bundle Adjustment -- Poster Session 8: Shape from X and Photometry -- Toward Efficient Acquisition of BRDFs with Fewer Samples -- Sato Shadow-Free TILT for Facade Rectification -- Reconstructing Shape from Dictionaries of Shading Primitives -- Iterative Feedback Estimation of Depth and Radiance from Defocused Images -- Two-Image Perspective Photometric Stereo Using Shape-from-Shading -- Stable Two View Reconstruction Using the Six-Point Algorithm -- Unknown Radial Distortion Centers in Multiple View Geometry Problems -- Depth-Estimation-Free Condition for Projective Factorization and Its Application to 3D Reconstruction -- Epipolar Geometry Estimation for Urban Scenes with Repetitive Structures -- Non-rigid Self-calibration of a Projective Camera -- Piecewise Planar Scene Reconstruction and Optimization for Multi-view Stereo -- A Bayesian Approach to Uncertainty-Based Depth Map Super Resolution -- Cross Image Inference Scheme for Stereo Matching -- Bayesian Epipolar Geometry Estimation from Tomographic Projections -- On the Global Self-calibration of Central Cameras Using Two Infinitesimal Rotations -- Adaptive Structure from Motion with a Contrario Model Estimation -- Precise 3D Reconstruction from a Single Image -- An Efficient Image Matching Method for Multi-View Stereo -- Self-calibration of a PTZ Camera Using New LMI Constraints -- Fast 3D Surface Reconstruction from Point Clouds Using Graph-Based Fronts Propagation -- Oral Session 9: Applications of Computer Vision Apparel Classification with Style -- Deblurring Vein Images and Removing Skin Wrinkle Patterns by Using Tri-band Illumination -- Reconstruction of 3D Surface and Restoration of Flat Document Image from Monocular Image Sequence -- Utilizing Optical Aberrations for Extended-Depth-of-Field Panoramas -- Poster Session 9: Low-level Vision and Applications of Computer Vision -- Motion-Invariant Coding Using a Programmable Aperture Camera -- Color-Aware Regularization for Gradient Domain Image Manipulation -- Local Covariance Filtering for Color Images -- A New Projection Space for Separation of Specular-Diffuse Reflection Components in Color Images -- Hand Vein Recognition Based on Oriented Gradient Maps and Local Feature Matching -- Fusing Warping, Cropping, and Scaling for Optimal Image Thumbnail Generation -- Shift-Map Based Stereo Image Retargeting with Disparity Adjustment -- Object Templates for Visual Place Categorization -- Reconstructing Sequential Patterns without Knowing Image Correspondences -- Registration of Multi-view Images of Planar Surfaces -- Automatic Stave Discovery for Musical Facsimiles -- Unsupervised Language Learning for Discovered Visual Concepts -- Parameterized Variety Based View Synthesis Scheme for Multi-view 3DTV -- Quasi-regular Facade Structure Extraction -- Multi-view Synthesis Based on Single View Reference Layer -- Hand-Eye Calibration without Hand Orientation Measurement Using Minimal Solution -- Detecting Changes in Images of Street Scenes -- Adaptive Background Defogging with Foreground Decremental Preconditioned Conjugate Gradient -- A Shadow Repair Approach for Kinect Depth Maps -- A Unified Framework for Line Extraction in Dioptric and Catadioptric Cameras -- Fusion of Time-of-Flight and Stereo for Disambiguation of Depth Measurements.