Copyright Philipp Slusallek Cs fall IBR: Model-based Methods Philipp Slusallek
Copyright Philipp Slusallek Cs fall Modeling and Rendering Traditional Pipeline: Modeling is hard Geometry: measurements, plans, user input Appearance: BRDF, texture Rendering is hard Complexity, reflection, lighting User inputModel: Geometry + MaterialImages ModelingRendering
Copyright Philipp Slusallek Cs fall Modeling and Computer Vision Computer Vision: Modeling from images Images contain geometric and appearance information Model Reconstruction is hard ImagesImage-based Model Calibration & Registration Accurate Model Images Model Reconstruction Rendering Image-based Rendering
Copyright Philipp Slusallek Cs fall Model Representation Representations: Geometry & Material Geometry & Textures Images with Depth (Range images, LDIs) Lightfield/Lumigraph Panorama Image-basedGeometry-based
Copyright Philipp Slusallek Cs fall Importance of Geometry
Copyright Philipp Slusallek Cs fall Image-based Rendering Advantages: Any geometry Photo-realistic: appearance is available Lower complexity Rendering is faster (?) Disadvantages: Sampled representation Visibility Data size Instability of CV algorithms
Copyright Philipp Slusallek Cs fall Model-based IBR Basic Idea: Sparse set of images [Debevec’97, Pulli’96] Overview: Approximate Modeling Photogrammetric modeling Triangulated depth maps View-dependent Texture Mapping Weighting Hardware accelerated rendering Model-based Stereo Details from stereo algorithms
Copyright Philipp Slusallek Cs fall Hybrid Approach Courtesy: P. Debevec
Copyright Philipp Slusallek Cs fall Approximate Modeling User-assisted photogrammetry [Debevec ‘97]: Based on “Structure from Motion” Use constraints in architectural models Approach: Simple block model Constraints reduce DOF Matching based on lines Non-linear optimization Initial Camera Positions
Copyright Philipp Slusallek Cs fall Approximate Modeling: Block Model Courtesy: P. Debevec
Copyright Philipp Slusallek Cs fall Approximate Modeling Active Light: Calibrated camera and projector Plane of light and triangulation Registration of multiple views Triangulation of point cloud Projector Camera
Copyright Philipp Slusallek Cs fall Approximate Modeling
Copyright Philipp Slusallek Cs fall Projecting Images Technique: Known camera positions Projective texture mapping Shadow buffer for occlusions Blending between textures Filling in
Copyright Philipp Slusallek Cs fall Visibility
Copyright Philipp Slusallek Cs fall Projecting Images
Copyright Philipp Slusallek Cs fall Projecting Images Simple compositing vs. blending Blending: Select “best” image closeness to viewing direction distance to border sampling density [Pulli] deletion of features Some computation in HW Smooth transition between pixels and frames Alpha blending, soft Z-buffer, confidence
Copyright Philipp Slusallek Cs fall Projecting Images Closeness to viewing direction: Triangulate the Hemisphere Delaunay triangulation of viewing directions Regular triangulation: label each vertex with best view Interpolate based on barycentric coordinates
Copyright Philipp Slusallek Cs fall Blending of Textures
Copyright Philipp Slusallek Cs fall Model-Based Stereo Problems with conventional stereo algorithms: Correspondences are difficult to find Large disparities Foreshortening, projective distortions Approach: Use approximate geometry to reproject one image Compute disparity of warped image Significant smaller disparity and foreshortening
Copyright Philipp Slusallek Cs fall Model-Based Stereo
Copyright Philipp Slusallek Cs fall Model-Based Stereo
Copyright Philipp Slusallek Cs fall Model-Based Stereo
Copyright Philipp Slusallek Cs fall Demos