|
I am interested in
problems related to computer graphics and
computer vision. The goal of my research is to develop novel sensing
methods and
algorithms for processing of visual information for scene analysis,
such as
extracting shape and intrinsic properties of objects, motion analysis
and
synthesizing realistic as well as computationally useful images. My
research is
targeted toward a diverse range of
applications such as motion photography, flash photography, 3D
reconstruction,
High dynamic range (HDR) imaging, image editing under variable
illumination, and
recovering intrinsic scene properties. Computational
Photography:
With advances in sensors and computational power, cameras coupled with a computer offer us new possibilities which were not possible with traditional cameras. Computational photography is emerging as a new field combining computer vision, graphics and photography to overcome the limitations of current cameras. Examples include combining multiple images taken with different camera parameters such as varying exposure and flash illumination, and camera arrays. At the same time, there is a need for novel sensors for specific applications that go beyond the traditional capturing of the scene as a regular grid of pixel intensities. My work in this area includes (a) combining images taken under flash illumination and without flash for removing flash artifacts and building flash-exposure HDR images for low-light indoor scenes, (b) a novel camera design for HDR imaging which measures static gradients instead of static intensities, and (c) a novel image capture system to encode motion blur in the scene, which can be subsequently deblurred easily and (d) mask-enhanced cameras for coded aperture and optical heterodyning Relevant Publications: A. Agrawal, R. Raskar, Shree K. Nayar & Y. Li, "Removing Photography Artifacts using Gradient Projection and Flash-Exposure Sampling", ACM SIGGRAPH 2005 J. Tumblin, A. Agrawal & R. Raskar, "Why I Want A Gradient Camera", IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2005 R. Raskar, A. Agrawal & J. Tumblin, "Coded Exposure Photography: Motion Deblurring using Fluttered Shutter", ACM SIGGRAPH 2006 A. Agrawal & R. Raskar, "Resolving Objects at Higher Resolution from a Single Motion-Blurred Image", IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2007 (oral presentation) A. Veeraraghavan, R. Raskar, A. Agrawal, A. Mohan & J. Tumblin, "Dappled Photography: Mask Enhanced Cameras for Heterodyned Light Fields and Coded Aperture Refocusing", ACM SIGGRAPH 2007 Surface Reconstruction from Gradient Fields: Reconstruction from gradient fields is the final step in several applications such as photometric stereo, shape from shading, retinex, mesh smoothing, HDR compression, image fusion, editing and image matting. In this work, I analyze the space of all possible reconstructions and propose a framework for obtaining meaningful solutions in this space. Previous solutions such as solving Poisson equation and Frankot-Chellappa algorithm are special cases of this framework. Relevant Publications: A. Agrawal, R. Raskar and R. Chellappa, "What is the Range of Surface Reconstructions from a Gradient Field?", European Conference on Computer Vision (ECCV), 2006 (oral presentation, 4.5% acceptance) pdf Matlab code GUI Matlab code Mfiles A. Agrawal, R. Chellappa & R. Raskar, "An Algebraic Approach to Surface Reconstruction from Gradient Fields", IEEE International Conference on Computer Vision (ICCV), 2005 pdf Matlab code Gradient Domain Algorithms for Scene Analysis: Gradient domain algorithms have recently become popular in computer graphics for applications such as image editing, matting, HDR compression, video surrealism and context enhancement. My focus in this work is to develop novel gradient based methods for "visiony" problems such as (a) recovering intrinsic image (separating illumination and reflectance) (b) removing shadows from color images (c) Edge suppression under significant illumination variations (d) removing glass reflections. Relevant Publications: A. Agrawal, R. Raskar and R. Chellappa, "Edge Suppression by Gradient Field Transformation Using Cross Projection Tensors", IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2006 pdf Matlab Code Previously, I worked on 3D Modeling and Visualization. |