A fully automatic method for descreening halftone images is presented based on convolutional neural networks with end-to-end learning. Incorporating context level information, the proposed method not only removes halftone artifacts but also synthesizes the fine details lost during halftone. The method consists of two main stages. In the first stage, intrinsic features of the scene are extracted, the low-frequency reconstruction of the image is estimated, and halftone patterns are removed. For the intrinsic features, the edges and object-categories are estimated and fed to the next stage as strong visual and contextual cues. In the second stage, fine details are synthesized on top of the low-frequency output based on an adversarial generative model. In addition, the novel problem of rescreening is addressed, where a natural input image is halftoned so as to be similar to a separately given reference halftone image. To this end, a two-stage convolutional neural network is also presented. Both networks are trained with millions of before-and-after example image pairs of various halftone styles. Qualitative and quantitative evaluations are provided, which demonstrates the effectiveness of the proposed methods.

Abstract Color quantization replaces the color of each pixel with the closest representative color, and thus it makes the resulting image partitioned into uniformly‐colored regions. As a consequence, continuous, detailed variations of color over the corresponding regions in the original image are lost through color quantization. In this paper, we present a novel blind scheme for restoring such variations from a color‐quantized input image without a priori knowledge of the quantization method. Our scheme identifies which pairs of uniformly‐colored regions in the input image should have continuous variations of color in the resulting image. Then, such regions are seamlessly stitched through optimization while preserving the closest representative colors. The user can optionally indicate which regions should be separated or stitched by scribbling constraint brushes across the regions. We demonstrate the effectiveness of our approach through diverse examples, such as photographs, cartoons, and artistic illustrations.

The need for the efficient handling of digital multimedia contents has increased tremendously as both home and mobile CE (consumer electronics) devices are equipped with large embedded storage such as hard disks or flash memories and are capable of handling enormous amounts of multimedia data. This paper presents Mint, an efficient multimedia contents management system for CE devices. It provides novel intuitive browsing and search interfaces by arranging multimedia data for various metadata. The system consists of the metadata extractors and the metadata core that are applicable to diverse hardware platforms and operating systems. Being applied to several CE devices has proved the usefulness of the system, especially for such devices as the digital TVs, set-top boxes and mobile phones

Abstract In this paper, we present an integrated framework of on‐line motion blending for locomotion generation. We first provide a novel scheme for incremental timewarping, which always guarantees that the time goes forward. Combining the idea of motion blending with that of posture rearrangement, we introduce a motion transition graph to address on‐line motion blending and transition simultaneously. Guided by a stream of motion specifications, our motion synthesis scheme moves from node to node in an on‐line manner while blending a motion at a node and generating a transition motion at an edge. For smooth on‐line motion transition, we also attach a set of example transition motions to an edge. To represent similar postures consistently, we exploit the inter‐frame coherency embedded in the input motion specification. Finally, we provide a comprehensive solution to on‐line motion retargeting by integrating existing techniques. Copyright © 2004 John Wiley & Sons, Ltd.

Real-time animation of human-like characters is an active research area in computer graphics. The conventional approaches have, however, hardly dealt with the rhythmic patterns of motions, which are essential in handling rhythmic motions such as dancing and locomotive motions. In this paper, we present a novel scheme for synthesizing a new motion from unlabelled example motions while preserving their rhythmic pattern. Our scheme first captures the motion beats from the example motions to extract the basic movements and their transitions. Based on those data, our scheme then constructs a movement transition graph that represents the example motions. Given an input sound signal, our scheme finally synthesizes a novel motion in an on-line manner while traversing the motion transition graph, which is synchronized with the input sound signal and also satisfies kinematic constraints given explicitly and implicitly. Through experiments, we have demonstrated that our scheme can effectively produce a variety of rhythmic motions.

Digital watermarking is one of commonly used solutions for copyright protection. A watermark should be imperceptible and robust to various attacks. In this paper, we address watermarking for motion data. Our watermarking scheme is based on two well‐known ideas, so called multiresolution representation and spread spectrum. We embed a watermark into a motion signal by perturbing large detail coefficients of its multiresolution representation, and extract the watermark by analyzing perturbation of coefficients from a suspected signal. For more effective watermark extraction, we align suspected motion data to the original using dynamic time warping. Our scheme has merits of spread spectrum such as the resilience to common signal processing as well as the robustness to time warping.