Youngsup Park

Motion Map Generation for Maintaining the Temporal Coherence of Brush Strokes

Painterly animation is a method that expresses images with a hand-painted appearance from a video, and the most crucial element for it is the coherence between frames. A motion map generation is proposed in this paper as a resolution to the issue of maintaining the coherence in the brush strokes between the frames. A motion map refers to the range of motion calculated by their magnitudes and directions between the frames with the edge of the previous frame as a point of reference. The different methods of motion estimation used in this paper include the optical flow method and the block-based method, and the method that yielded the biggest PSNR using the motion information (the directions and magnitudes) acquired by various methods of motion estimation has been chosen as the final motion information to form a motion map. The created motion map determined the part of the frame that should be re-painted. In order to maintain the temporal coherence, the motion information was applied to only the strong edges that determine the directions of the brush strokes. Also, this paper sought to reduce the flickering phenomenon between the frames by using the multiple exposure method and the difference map created by the difference between images of the source and the canvas. Maintenance of the coherence in the direction of the brush strokes was also attempted by a local gradient interpolation in an attempt to maintain the structural coherence.

Rendering Technique for Colored Paper Mosaic

The work presented in this paper shows the way to generate colored paper mosaics using computer graphics techniques. Following two tasks need to be done to generate colored paper mosaic. The first one is to generate colored paper tile and the other one is to arrange the tile. Voronoi Diagram and Random Point Displacement have been used in this paper to come up with the shape of the tile. And, energy value that the tile has depending on its location is the factor to determine the best positioning of the tile. This paper focuses on representing the overlap among tiles, maintenance of the edge of input image, and various shapes of tiles in the final output image by solving two tasks mentioned above.

BRDF recovering and scene re-lighting using an environment map

A rendering algorithm based on physical phenomenon and well-defined model data are needed to generate a synthetic image realistically. In this paper, inverse rendering based on an environment map is proposed to obtain BRDF of objects. Many researchers try to find BRDF by analyzing many pictures taken with well-controlled camera and light position. But this method is available at a laboratory environment. Limits of previous works can be conquered by using inverse rendering based on an environment map and it is possible to get BRDF of objects at a general environment. Inputs are an image, an environment map and 3D geometry data of models. Objects of an input image are divided into three parts - diffuse reflection, glossy reflection and diffuse texture -and BRDF of objects are recovered. Using BRDF of objects, it is possible to add virtual objects and to remove real objects from an input image.

Dynamic brush stroke generation for an impressionist effect

We propose dynamic brush stroke generation for an impressionist effect of source images, using reference data. Colors used are formed by actual palette colors from artists. To create the palette, we have referred mostly to colors used in Van Gogh’s works and determined the color of brush strokes by transferring it to the most similar one, through comparing colors used in source images and the palette colors. Also, by referring to the edge orientation of source images, we have applied a brush stroke direction that surrounds the edges. The sizes were determined depending on the different sizes of the objects from wide to narrow brushes. Finally, we applied spline curve and line shapes. The brush strokes created in such method, were applied separately according to its segmented images, and composed after rendering.

View morphing using sprites with depth

Recently there has been a great deal of interest in morphing techniques to produce intermediary images between two images. These techniques combine 2D interpolations of shape and color to make realistic special effects. The authors simulate view morphing, using sprites with depth about a pair of partially occluded images based on the epipolar geometry. The epipolar constraints improve the accuracy of the estimated disparity maps. To compute a depth map between two images, we use use a simple stereo system. Depth can be estimated from the disparity of corresponding points and is inversely proportional to disparity, used when sprites are composed. A sprite is an image with the standard three-color components (red, green, blue) and an additional channel that encodes the images shape. In the case of a pair of partially occluded images, whether there should be decomposed sprites to an image is the reason for attributing to reduce two problems (folds and holes), in the view morphing procedures.