Lee Markosian

Real-time nonphotorealistic rendering

Nonphotorealistic rendering (NPR) can help make comprehensible but simple pictures of complicated objects by employing an economy of line. But current nonphotorealistic rendering is primarily a batch process. This paper presents a real-time nonphotorealistic renderer that deliberately trades accuracy and detail for speed. Our renderer uses a method for determining visible lines and surfaces which is a modification of Appel’s hidden-line algorithm, with improvements which are based on the topology of singular maps of a surface into the plane. The method we describe for determining visibility has the potential to be used in any NPR system that requires a description of visible lines or surfaces in the scene. The major contribution of this paper is thus to describe a tool which can significantly improve the performance of these systems. We demonstrate the system with several nonphotorealistic rendering styles, all of which operate on complex models at interactive frame rates. CR

Artistic silhouettes: a hybrid approach

We present a new algorithm for rendering silhouette outlines of 3D polygonal meshes with stylized strokes. Rather than use silhouette edges of the model directly as the basis for drawing strokes, we first process the edges in image space to create long, connected paths corresponding to visible portions of silhouettes. The resulting paths have the precision of object-space edges, but avoid the unwanted zig-zagging and inconsistent visibility of raw silhouette edges. Our hybrid screen/object space approach thus allows us to apply stylizations to strokes that follow the visual silhouettes of an object. We describe details of our OpenGL-based stylized strokes that can resemble natural media, but render at interactive rates. We demonstrate our technique with the accompanying still images and animations rendered with our technique. CR Categories and Subject Descriptors: I.3.3 [Computer Graphics]: Picture/Image Generation - Display algorithms

Art-based rendering of fur, grass, and trees

Artists and illustrators can evoke the complexity of fur or vegetation with relatively few well-placed strokes. We present an algorithm that uses strokes to render 3D computer graphics scenes in a stylized manner suggesting the complexity of the scene without representing it explicitly. The basic algorithm is customizable to produce a range of effects including fur, grass and trees, as we demonstrate in this paper and accompanying video. The algorithm is implemented within a broader framework that supports procedural stroke-based textures on polyhedral models. It renders moderately complex scenes at multiple frames per second on current graphics workstations, and provides some interframe coherence. CR

An interface for sketching 3D curves

The ability to specify nonplanar 3D curves is of fundamental importance in 3D modeling and animation systems. Effective techniques for specifying such curves using 2D input devices are desirable, but existing methods typically require the user to edit the curve from several viewpoints. We present a novel method for specifying 3D curves with 2D input from a single viewpoint. The user rst draws the curve as it appears from the current viewpoint, and then draws its shadow on the oor plane. The system correlates the curve with its shadow to compute the curves 3D shape. This method is more natural than existing methods in that it leverages skills that many artists and designers have developed from work with pencil and paper.

Line drawings via abstracted shading

We describe a GPU-based algorithm for rendering a 3D model as a line drawing, based on the insight that a line drawing can be understood as an abstraction of a shaded image. We thus render lines along tone boundaries or thin dark areas in the shaded image. We extend this notion to the dual: we render highlight lines along thin bright areas and tone boundaries. We combine the lines with toon shading to capture broad regions of tone. The resulting line drawings effectively convey both shape and material cues. The lines produced by the method can include silhouettes. creases, and ridges, along with a generalization of suggestive contours that responds to lighting as well as viewing changes. The method supports automatic level of abstraction, where the size of depicted shape features adjusts appropriately as the camera zooms in or out. Animated models can be rendered in real time because costly mesh curvature calculations are not needed.

Skin: a constructive approach to modeling free-form shapes

We present a new particle-based surface representation with which a user can interactively sculpt free-form surfaces. The particles maintain mesh connectivity and operate under rules that lead them to form triangulations with properties that make them suitable for use in subdivision. A user interactively guides the particles, which we call skin, to grow over a given collection of polyhedral elements (or skeletons), yielding a smooth surface (through subdivision) that approximates the underlying skeletal shapes. Skin resembles blobby modeling in the constructive approach to modeling it supports, but allows a richer vocabulary of skeleton shapes, supports sharp creases where desired, and provides a convenient mechanism for adding multiresolution surface detail. CR

Art-based rendering with continuous levels of detail

In previous work [6], we presented an algorithm for rendering virtual scenes using art-based styles. We demonstrated the ability to render fur, grass, and trees in a stylized manner that evoked the complexity of these textures without representing all their components explicitly. We achieved this with stroke-based procedural textures that generated detail elements, or graftals, just as needed. Our implementation had several drawbacks. First, each new graftal texture required a procedural implementation that included writing code. Also, graftals were regenerated in each frame in a way that led to excessive introduction and elimination of graftals even for small changes in camera parameters. Lastly, our system provided no way to continuously vary the properties of graftals, including color, size, or stroke width. Such an ability could be used to achieve better frame-to-frame coherence, or more generally to animate graftals. In this paper, we present a new framework for graftal textures that addresses these issues. Our new framework allows all major decisions about graftal look and behavior to be specified in a text file that can be edited by a designer. We have achieved greater frameto-frame coherence by using graftals that remain in fixed positions on the model surface. The look and behavior of graftals as they appear or disappear can now be animated to create smooth transitions. Finally, we introduce the concept of tufts which manage the multiresolution behavior of graftals according to the specifications of the scene designer.

Stroke Pattern Analysis and Synthesis

We present a synthesis technique that can automatically generate stroke patterns based on a user‐specified reference pattern. Our method is an extension of texture synthesis techniques to vector‐based patterns. Such an extension requires (a) an analysis of the pattern properties to extract meaningful pattern elements (defined as clusters of strokes) and (b) a synthesis algorithm based on similarities in the detected stroke clusters. Our method is based on results from human vision research concerning perceptual organization. The resulting synthesized patterns effectively reproduce the properties of the input patterns, and can be used to fill both 1D paths and 2D regions.

Seamless interaction in virtual reality

Jot is a novel research interface for virtual reality modeling. This system seamlessly integrates and applies a variety of virtual and physical tools, each customized for specific tasks. The Jot interface not only moves smoothly from one tool to another but also physically and cognitively matches individual tools to the tasks they perform. In particular, we exploit the notion that gestural interaction is more direct, in many cases, than traditional widget based interaction. We also respect the time tested observation that some operations-even conceptually three dimensional ones-are better performed with 1D or 2D input devices, whereas other operations are more naturally performed using stereoscopic views, higher DOF input devices, or both. Ultimately we strive for a 3D modeling system with an interface as transparent as the interaction afforded by a pencil and a sheet of paper. For example, the system should facilitate the tasks of drawing and erasing and provide an easy transition between the two. Jot emerged from our previous work on a mouse based system, called Sketch, for gesturally creating imprecise 3D models. Jot extends Sketchs functionality to a wider spectrum of modeling, from concept design to detailed feature based parametric parts. Jot also extends the interaction in Sketch to better support individual modeling tasks. We extended Sketchs gestural framework to integrate interface components ranging from traditional desktop interface widgets to context sensitive gestures to direct manipulation techniques originally designed for immersive VR.

Dynamic 2D patterns for shading 3D scenes

We describe a new way to render 3D scenes in a variety of non-photorealistic styles, based on patterns whose structure and motion are defined in 2D. In doing so, we sacrifice the ability of patterns that wrap onto 3D surfaces to convey shape through their structure and motion. In return, we gain several advantages, chiefly that 2D patterns are more visually abstract - a quality often sought by artists, which explains their widespread use in hand-drawn images. Extending such styles to 3D graphics presents a challenge: how should a 2D pattern move? Our solution is to transform it each frame by a 2D similarity transform that closely follows the underlying 3D shape. The resulting motion is often surprisingly effective, and has a striking cartoon quality that matches the visual style.