Junhwan Kim - Academia.edu (original) (raw)
Uploads
Papers by Junhwan Kim
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
We address visual correspondence problems without assumingthat scene points have similar intensit... more We address visual correspondence problems without assumingthat scene points have similar intensities in differentviews. This situation is common, usually due tonon-lambertian scenes or to di# erences between cameras. We use maximization of mutual information, apowerful technique for registering images that requiresno apriorimodel of the relationship between scene intensitiesin di# erent views. However, it has provendi# cult to use mutual information to compute densevisual correspondence. Comparing fixed-size windowsvia ...
Bookmarks Related papers MentionsView impact
Proceedings Ninth IEEE International Conference on Computer Vision, 2003
Bookmarks Related papers MentionsView impact
Lecture Notes in Computer Science, 2002
Bookmarks Related papers MentionsView impact
Journal of X-ray science and technology, 2003
Medical imaging often involves the injection of contrast agents and the subsequent analysis of ti... more Medical imaging often involves the injection of contrast agents and the subsequent analysis of tissue enhancement patterns. Many important types of tissue have characteristic enhancement patterns; for example, in MR mammography, malignancies exhibit a characteristic "wash out" temporal pattern, while in MR angiography, arteries, veins and parenchyma each have their own distinctive temporal signature. In such time resolved image series, there are substantial changes in intensities; however, this change is due primarily to the contrast agent, rather than to motion. As a result, the task of automatically segmenting contrast-enhanced images poses interesting new challenges. In this paper, we propose a new image segmentation algorithm for time resolved image series with contrast enhancement, using a model-based time series analysis of individual pixels. We take an energy minimization approach to ensure spatial coherence. The energy is minimized in an expectation-maximization fa...
Bookmarks Related papers MentionsView impact
Magnetic Resonance in Medicine, 2002
Bookmarks Related papers MentionsView impact
ACM Transactions on Graphics, 2002
This paper introduces a new kind of mosaic, called Jigsaw Image Mosaic (JIM), where image tiles o... more This paper introduces a new kind of mosaic, called Jigsaw Image Mosaic (JIM), where image tiles of arbitrary shape are used to compose the final picture. The generation of a Jigsaw Image Mosaic is a solution to the following problem: given an arbitrarily-shaped container image and a set of arbitrarily-shaped image tiles, fill the container as compactly as possible with tiles of similar color to the container taken from the input set while optionally deforming them slightly to achieve a more visually-pleasing effect. We approach the problem by defining a mosaic as the tile configuration that minimizes a mosaicing energy function. We introduce a general energy-based framework for mosaicing problems that extends some of the existing algorithms such as Photomosaics and Simulated Decorative Mosaics. We also present a fast algorithm to solve the mosaicing problem at an acceptable computational cost. We demonstrate the use of our method by applying it to a wide range of container images an...
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
We address visual correspondence problems without assumingthat scene points have similar intensit... more We address visual correspondence problems without assumingthat scene points have similar intensities in differentviews. This situation is common, usually due tonon-lambertian scenes or to di# erences between cameras. We use maximization of mutual information, apowerful technique for registering images that requiresno apriorimodel of the relationship between scene intensitiesin di# erent views. However, it has provendi# cult to use mutual information to compute densevisual correspondence. Comparing fixed-size windowsvia ...
Bookmarks Related papers MentionsView impact
Proceedings Ninth IEEE International Conference on Computer Vision, 2003
Bookmarks Related papers MentionsView impact
Lecture Notes in Computer Science, 2002
Bookmarks Related papers MentionsView impact
Journal of X-ray science and technology, 2003
Medical imaging often involves the injection of contrast agents and the subsequent analysis of ti... more Medical imaging often involves the injection of contrast agents and the subsequent analysis of tissue enhancement patterns. Many important types of tissue have characteristic enhancement patterns; for example, in MR mammography, malignancies exhibit a characteristic "wash out" temporal pattern, while in MR angiography, arteries, veins and parenchyma each have their own distinctive temporal signature. In such time resolved image series, there are substantial changes in intensities; however, this change is due primarily to the contrast agent, rather than to motion. As a result, the task of automatically segmenting contrast-enhanced images poses interesting new challenges. In this paper, we propose a new image segmentation algorithm for time resolved image series with contrast enhancement, using a model-based time series analysis of individual pixels. We take an energy minimization approach to ensure spatial coherence. The energy is minimized in an expectation-maximization fa...
Bookmarks Related papers MentionsView impact
Magnetic Resonance in Medicine, 2002
Bookmarks Related papers MentionsView impact
ACM Transactions on Graphics, 2002
This paper introduces a new kind of mosaic, called Jigsaw Image Mosaic (JIM), where image tiles o... more This paper introduces a new kind of mosaic, called Jigsaw Image Mosaic (JIM), where image tiles of arbitrary shape are used to compose the final picture. The generation of a Jigsaw Image Mosaic is a solution to the following problem: given an arbitrarily-shaped container image and a set of arbitrarily-shaped image tiles, fill the container as compactly as possible with tiles of similar color to the container taken from the input set while optionally deforming them slightly to achieve a more visually-pleasing effect. We approach the problem by defining a mosaic as the tile configuration that minimizes a mosaicing energy function. We introduce a general energy-based framework for mosaicing problems that extends some of the existing algorithms such as Photomosaics and Simulated Decorative Mosaics. We also present a fast algorithm to solve the mosaicing problem at an acceptable computational cost. We demonstrate the use of our method by applying it to a wide range of container images an...
Bookmarks Related papers MentionsView impact