Image conversion

About: Image conversion is a research topic. Over the lifetime, 2490 publications have been published within this topic receiving 19077 citations.

Method and apparatus for optimizing and storing contone images for subsequent half-toning and merging with text

Abstract: In publishing listings of real estate properties, photographed (or videotaped) images are converted to digital graphics at the "front end" of the publishing process. This image conversion ("capture") process includes cropping, contrast adjustment using statistical techniques, and generation of control information needed later in the process. A library of digital graphics and associated information is maintained. Graphics selected from this library and a conventional "multiple listing service" text database are then merged during a text composition process to provide a stream of digital data including text and embedded graphics to be printed in the listing book. Because all image operations (e.g., sizing, cropping, and digital image quality enhancement) are performed when the images are captured, no time consuming post processing steps are requird. Image capture is performed in a single step, and once an image is captured it can be used "as is" in the printing process without further review, modification or post processing.

Automatic 2D-to-3D image conversion using 3D examples from the internet

Abstract: The availability of 3D hardware has so far outpaced the production of 3D content. Although to date many methods have been proposed to convert 2D images to 3D stereopairs, the most successful ones involve human operators and, therefore, are time-consuming and costly, while the fully-automatic ones have not yet achieved the same level of quality. This subpar performance is due to the fact that automatic methods usually rely on assumptions about the captured 3D scene that are often violated in practice. In this paper, we explore a radically different approach inspired by our work on saliency detection in images. Instead of relying on a deterministic scene model for the input 2D image, we propose to "learn" the model from a large dictionary of stereopairs, such as YouTube 3D. Our new approach is built upon a key observation and an assumption. The key observation is that among millions of stereopairs available on-line, there likely exist many stereopairs whose 3D content matches that of the 2D input (query). We assume that two stereopairs whose left images are photometrically similar are likely to have similar disparity fields. Our approach first finds a number of on-line stereopairs whose left image is a close photometric match to the 2D query and then extracts depth information from these stereopairs. Since disparities for the selected stereopairs differ due to differences in underlying image content, level of noise, distortions, etc., we combine them by using the median. We apply the resulting median disparity field to the 2D query to obtain the corresponding right image, while handling occlusions and newly-exposed areas in the usual way. We have applied our method in two scenarios. First, we used YouTube 3D videos in search of the most similar frames. Then, we repeated the experiments on a small, but carefully-selected, dictionary of stereopairs closely matching the query. This, to a degree, emulates the results one would expect from the use of an extremely large 3D repository. While far from perfect, the presented results demonstrate that on-line repositories of 3D content can be used for effective 2D-to-3D image conversion. With the continuously increasing amount of 3D data on-line and with the rapidly growing computing power in the cloud, the proposed framework seems a promising alternative to operator-assisted 2D-to-3D conversion.

Pixel quantization with adaptive error diffusion

Abstract: A method of image conversion takes an original print ready image at a first resolution scaling and orientation, and simulates printing that image by creating a representation of a page at the particular resolution, scaling and orientation. The image is then periodically sampled through the page with an aperture that corresponds to the desired output. Because the resolution, scaling and/or orientation of the "print" and "scan" are distinct, the aperture "sees" areas which may correspond to more than a single pixel on the original image, and thereby may derive a signal that is gray, i.e., the aperture may see an area that is neither completely black or white, and the image data derived will be considered gray. The gray image data, which may be definable at several bits per pixel, is then made print-ready by reducing the number of bits defining each pixel to a valid output state. The reduction step is accomplished through an adaptive error diffusion method, in which thresholding error is directed only to members of a set of error receiving pixels already having a valid output state. When all the error receiving pixels have a valid output state, error is directed to those pixels in accordance with a standard error diffusion method.

Method for making image conversions with error diffusion

Abstract: A method of image conversion takes an original print ready image at a first resolution scaling and orientation, and simulates printing that image by creating a representation of a page at the particular resolution, scaling and orientation. The image is then periodically sampled through the page with an aperture that corresponds to the desired output. Because the resolution, scaling and/or orientation of the "print" and "scan" are distinct, the aperture "sees" areas which may correspond to more than a single pixel on the original image, and thereby may derive a signal that is gray, i.e., the aperture may see an area that is neither completely black or white, and the image data derived will be considered gray. The gray image data, which may be definable at several bits per pixel, is then made print-ready by reducing the number of bits defining each pixel to a valid output state (0, 1 for a typical binary printer, 0, 1, 2, 3 for a quaternary printer, etc.). The reduction step is accomplished through error diffusion methods that maintain the local area gray density level. In color applications, each of the bitmaps representing a color separation, be it 2, 3, 4 or any other number, is treated independently with the same method.

New television with 2D/3D image conversion technologies

Abstract: The image conversion technologies from 2D images into 3D images with the `Modified Time Difference' method are proposed. These technologies allow to convert automatic and real-time ordinary 2D images into binocular parallax 3D images according to the detected movements of objects in the images. We integrated circuits such as a movement detector, a delay time controller and a delay direction controller into a single LSI chip to make a 2D/3D conversion board compact. We put this conversion board into a television set to introduce a new type of 3D consumer television with which we all can enjoy converted 3D images originally provided by TVs, VCRs and the like. Vertical frequency of this 3D television is 120 Hz and twice as fast as the ordinary television for providing flickerless 3D image.