Publications

We present a procedural technique for the controllable synthesis of detailed terrains. We generate terrains based on a sparse curve network representation, where interconnected curves are distributed in the plane and can be procedurally assigned height. We employ path planning to procedurally generate irregular curves around user-designated peaks. Optionally, the user can specify base signals for the curves. Then we assign height to the curves using random walks with controlled probability distributions, a process which can produce signals with a variety of shapes. The curve network partitions space into individual patches. We interpolate patch heights using mean value coordinates, after which we have a complete terrain heightfield. Our algorithm enables user to obtain prominent features with lightweight interaction. Increasing the density of curves and roughness of curve profiles adds detail to the synthetic terrains. The curves in a network are organized into a hierarchy, where the major curves are created first and the curves constructed at later stages are affected by earlier curves. Our approach is capable of producing a variety of landscapes with prominent ridges and distinct shapes

@inproceedings{ariyan15terrain,
    author = {Maryam Ariyan and David Mould},
    title = {Terrain synthesis using curve networks},
    booktitle = {Graphics Interface},
    year = 2015,
}
                    

Encouraging middle school girls to become interested in Computer Science has been a concern in the Computer Science Education community for many years. Although many girls do play digital games[4], this has not led to further interest in computer science. We believe this is caused by many factors. We are researching the effect of teaching (non-programming) computer science principles embedded in after school games, to encourage girls to study computer science in their future. We maintain that the experience of learning these principles will increase their confidence that they can enjoy learning more about computer science. This on-going NSF sponsored AISL research project is described for discussion.

@inproceedings{stewart15girls,
    author = {Carolee Stewart-Gardiner and Gail Carmichael and Elisabeth Gee and Lorraine Hopping},
    title = {Girls Learning Computer Science Principles with After School Games},
    booktitle = {GenderIT},
    year = 2015,
}
                    

In this study, we are investigating the role of story in promoting girls’ engagement with and learning from games that aim to foster understanding of computer science concepts. Story is commonly cited as an element of games that might have particular appeal for girls (Kelleher, Pausch, and Kiesler, 2007). While we are skeptical of broad generalizations about girls’ or boys’ preferences for particular game features, we believe that stories have affordances for engagement and learning that might help to overcome, for example, stereotypes or misconceptions about domains such as computer science. Here we describe the initial stages of our research, including an analysis of the role of story in existing STEM games and the set of analog game prototypes that we have created to introduce several core concepts of computer science.

@inproceedings{gee15story,
    author = {Elisabeth Gee and Kelly Tran and Carolee Stewart-Gardiner and Gail Carmichael and Lorraine Hopping},
    title = {The role of story in computer science games for girls},
    booktitle = {Games, Learning and Society Conference},
    year = 2015,
}
                    

We propose a two-stage approach to painterly rendering of photographs, where the image plane is first warped to produce a distorted or caricatured effect and then the resulting image is rendered with a painterly effect. We use SLIC superpixels to obtain an oversegmentation, and assign spring parameters uniformly to all pixels within a region; then, the mass-spring simulation distorts the plane in a random but content-sensitive way. With aggressive warping, the subsequent painterly rendering can be done lightly and need not remove much detail. The resulting renderings convey a sense of being painted and leave a sense of being handmade and not overly beholden to the photographic scene.

@inproceedings{li15warping,
    author = {Jiayu Li and David Mould},
    title = {Image Warping for a Painterly Effect},
    booktitle = {Expressive},
    year = 2015,
}
                    

We propose guiding vectors to augment graph-based tree synthesis, in which trees are collections of least-cost paths in a graph. Each node has an associated guiding vector; edges parallel to the guiding vector are cheap, but edges are more expensive when their orientation differs from the guiding vector.We further propose an incremental method for assigning guiding vectors over the graph, in which a node's guiding vector is an incremental rotation of that of its parent. We present a complete procedural system for tree modeling; our use of guiding vectors enables the graph-based method to produce high-quality tree models resembling a variety of real-world tree species.

@journal{xu15guided,
    author = {Ling Xu and David Mould},
    title = {Procedural Tree Modeling with Guiding Vectors},
    booktitle = {Computer Graphics Forum},
    year = 2015,
}
                    

We introduce a contextual descriptor which aims to provide a geometric description of the functionality of a 3D object in the context of a given scene. Differently from previous works, we do not regard functionality as an abstract label or represent it implicitly through an agent. Our descriptor, called interaction context or ICON for short, explicitly represents the geometry of object-to-object interactions. Our approach to object functionality analysis is based on the key premise that functionality should mainly be derived from interactions between objects and not objects in isolation. Specifically, ICON collects geometric and structural features to encode interactions between a central object in a 3D scene and its surrounding objects. These interactions are then grouped based on feature similarity, leading to a hierarchical structure. By focusing on interactions and their organization, ICON is insensitive to the numbers of objects that appear in a scene, the specific disposition of objects around the central object, or the objects' fine-grained geometry. With a series of experiments, we demonstrate the potential of ICON in functionality-oriented shape processing, including shape retrieval (either directly or by complementing existing shape descriptors), segmentation, and synthesis.

@article{hu15icon,
    author = {Ruizhen Hu and Chenyang Zhu and Oliver van Kaick and Ligang Liu and Ariel Shamir and Hao Zhang},
    title = {Interaction Context (ICON): Towards a Geometric Functionality Descriptor},
    journal = {ACM Trans. on Graphics (Proc. SIGGRAPH)},
    volume = {34},
    number = {4},
    pages = {83:1--83:12},
    year = 2015,
}
                    

Computing similarities or distances between 3D shapes is a crucial building block for numerous tasks, including shape retrieval, exploration and classification. Current state-of-the-art distance measures mostly consider the overall appearance of the shapes and are less sensitive to fine changes in shape structure or geometry. We present shape edit distance (SHED) that measures the amount of effort needed to transform one shape into the other, in terms of rearranging the parts of one shape to match the parts of the other shape, as well as possibly adding and removing parts. The shape edit distance takes into account both the similarity of the overall shape structure and the similarity of individual parts of the shapes. We show that SHED is favorable to state-of-the-art distance measures in a variety of applications and datasets, and is especially successful in scenarios where detecting fine details of the shapes is important, such as shape retrieval and exploration.

@article{kleiman15shed,
    author = {Yanir Kleiman and Oliver van Kaick and Olga Sorkine-Hornung and Daniel Cohen-Or},
    title = {SHED: Shape Edit Distance for Fine-grained Shape Similarity},
    journal = {ACM Trans. on Graphics (Proc. SIGGRAPH Asia)},
    volume = {34},
    number = {6},
    pages = {235:1--235:11},
    year = 2015,
}
                    

This paper investigates contrast enhancement as an approach to tone reduction, aiming to convert a photograph to black and white. Using a filter-based approach to strengthen contrast, we avoid making a hard decision about how to assign tones to segmented regions. Our method is inspired by sticks filtering, used to enhance medical images but not previously used in non-photorealistic rendering. We amplify contrast of pixels along the direction of greatest local difference from the mean, strengthening even weak features if they are most prominent. A final thresholding step converts the contrast-enhanced image to black and white. Local smoothing and contrast enhancement balances abstraction and structure preservation; the main advantage of our method is its faithful depiction of image detail. Our method can create a set of effects: line drawing, hatching, and black and white, all having superior details to previous black and white methods.

@article{li15bwimages,
    author = {Hua Li and David Mould},
    title = {Contrast-enhanced black and white images},
    journal = {Computer Graphics Forum},
    year = 2015,
}
                    

The rise of game development and game studies on university campuses prompts academic libraries to consider how to support teaching and research in this area. This article examines current issues and challenges in the development of game collections at academic libraries. The gaming ecosystem has become more complex and libraries may need to move beyond collections largely based on console video games. This article will advance the discussion by considering emerging issues to support access to the full range of games. The article will use examples from Carleton University Library, Ottawa, which has been developing a game collection since 2008.

@article{cross,
    author = {Emma Cross and David Mould and Robert Smith},
    title = {The protean challenge of game collections at academic libraries},
    journal = {New Review of Academic Librarianship},
    year = 2015,
}
                    

Tree models are widely needed in 3D movies, games, and artistic presentations. Their styles abound. Due to the complexity and the variety of trees, realistic tree modeling remains a challenge. For automated methods, the objectives of speed and controllability are elusive. Data-driven and manual approaches have their own drawbacks: it is inconvenient to obtain laser scans or photographs of a desired tree, and human intervention can be costly. Another different style of trees, abstract trees, possesses stylized branches and aesthetic patterns. However, abstract tree modeling with an emphasis on branching structures receives less attention. Abstract trees in the art style of Gustav Klimt has never been explored in computer graphics.

@thesis{xu15thesis,
    author = {Ling Xu},
    title = {Abstract and Realistic Tree Modeling},
    booktitle = {Carleton University},
    year = 2015,
}
                    

Partition of unity parametrics (PUPs) are a recent framework designed for geometric modeling. We propose employing PUPs for procedural texture synthesis, taking advantage of the framework's guarantees of high continuity and local support. Using PUPs to interpolate among data values distributed through the plane, the problem of texture synthesis can be approached from the perspective of point placement and attribute assignment; a graph over the points, such as the Delaunay triangulation, allows us to apply attributes or labels to the nodes in a structured way. We present several alternative mechanisms for point distribution and demonstrate how the system is able to produce a variety of distinct classes of texture, including analogs to cellular texture, Perlin noise, and progressively-variant textures. We further describe structured textures obtained by assigning label clusters using queries over the graph, such as breadth-first or depth-first traversal.

@article{PUPs,
    author = {Jacquelin Caron and David Mould},
    title = {Texture Synthesis Using Label Assignment Over A Graph},
    journal = {Computers and Graphics},
    year = 2014,
}
                    

Crafting satisfying narratives while preserving player freedom is a longstanding challenge for computer games. The quest structure used by many games allows players to experience content nonlinearly, but risks creating disjointed stories when side quests only minimally integrate with the main story. Instead, we propose a flexible, scene-based emergent story system that reacts to the player's actions while maintaining a reasonable amount of authorial control over the story. Based on the philosophy of story scenes as kernels or satellites, we define a minimal story graph that initially contains mostly disconnected nodes. Over time, the graph is built dynamically from the most reasonable choices offered to the player. In this paper, we describe the framework of our system and present an early prototype game as a case study. We end with a vision of how our framework could be used to create more coherent, emergent stories in games.

@inproceedings{Carmichael:14:AFC,
    author = {Gail Carmichael and David Mould},
    title = {A Framework for Coherent Emergent Stories},
    booktitle = {Foundations of Digital Games},
    year = 2014,
}
                    

Although stories in games have become more sophisticated over time, their use of nonlinear techniques has not yet become as prevalent as in traditional media like novels and films. Writers have largely excluded nonlinear techniques from their toolbox, possibly because of fears of introducing inconsistencies when player actions alter past events. However, as we show through a survey of common nonlinear techniques seen in television, novels, and film, games can and have avoided these inconsistencies while maintaining gameplay agency. Many players prefer a high quality static story incorporated into strong gameplay, making the insight from this discussion immediately useful in designing nonlinear game stories. We also discuss future directions for how nonlinear techniques can offer both gameplay and story agency, hopefully bringing the quality of game stories one step closer to their traditional counterparts.

@inproceedings{carmichael14stories,
    author = {Gail Carmichael and David Mould},
    title = {Chronologically Nonlinear Techniques in Traditional Media and Games},
    booktitle = {Foundations of Digital Games (WIP)},
    year = 2014,
}
                    

Non-photorealistic rendering (NPR) algorithms are used to produce stylized images, e.g., in a painted or stippled style. This thesis presents a new family of automatic methods for effects and styles including halftoning, screening, stippling, line art, and mosaics. Our proposed algorithms are motivated by perceptual effects including contrast and similarity to present structure preservation. Meanwhile, the novelty also shows in exploring primitives in distribution for our NPR algorithms.

@thesis{li14thesis,
    author = {Hua Li},
    title = {Perception-Motivated High Quality Stylization},
    booktitle = {Carleton University},
    year = 2014,
}
                    

Our paper provides an approach to create line-drawing stylizations of input images. The main idea is to use particle tracing with interaction between nearby particles: the particles coordinate their movements so as to produce varied but roughly parallel traces. The particle density varies according to the tone in the input images, thereby expressing bright and dark areas. Using procedural distributions of particles, we can also generate smooth abstract patterns.

@inproceedings{wei14particle,
    author = {Chujia Wei and David Mould},
    title = {Coordinated Particle Systems for Image Stylization},
    booktitle = {Graphics Interface},
    year = 2014,
}
                    

This paper presents a design-focused analysis of 36 educational games in the domain of computer science. Given the importance of broadening participation in computing and especially improving gender diversity, it takes a gender-inclusive view on the design of such games, examining design features that are related to gender, storytelling, and fostering an inclusive environment. Ten broad patterns emerge from applying the analytical framework, such as the prevalence of robots, the portrayal of computing as individualistic, and the dominance of puzzle elements. The framework and resulting patterns can be used to inspire the future development of educational games teaching computer science, through identification of new opportunities, as well as to guide the evaluation of their design.

@inproceedings{harteveld14design,
    author = {Casper Harteveld and Gillian Smith and Gail Carmichael and Elisabeth Gee and Carolee Stewart-Gardiner},
    title = {A Design-Focused Analysis of Games Teaching Computer Science.},
    booktitle = {Games and Learning Society Conference},
    year = 2014,
}
                    

I argue in favor of a systematic subjective evaluation of non-photorealistic images. Objective measurements are hard to design, and quantitative user studies are problematic for a multiplicity of reasons. Subjective evaluations are not quantitative but are faster to conduct and offer the chance to dig into subtleties that are obscured by numerical scores. By carefully laying out the important elements of the intended image style, and then evaluating their results according to their adherence to the style, researchers can produce convincing evaluations with a manageable level of effort.

@inproceedings{mould14NPR,
    author = {David Mould},
    title = {Authorial Subjective Evaluation of Non-Photorealistic Images},
    booktitle = {NPAR},
    year = 2014,
}
                    

The shortage of females in computer science has been studied before. Computer games have long been one way teenage boys find an interest in Computer Science, but most of those games are not appealing to teenage girls. This paper describes the ongoing collaborative research project which is experimenting with the design of educational computer games. Our research has the objective to influence middle school girls to pursue computer science in high school and college. The games are designed to change the image of computing among middle school girls, and to instill confidence by teaching real computer science concepts through puzzles. Gail Carmichael and her team of graduate students at Carleton University designed and created an educational computer game (Grams House) in 2010 with a "helping others" story. The prototype game focuses on two computer concept puzzles. In summer 2012, two undergraduates Jennifer Latham, and Nathaly Lozano, at Kean University designed and created a companion game (Grams Grocery Shop) with more teen appeal , and two more puzzles. In fall 2012 the Kean University research team piloted the game pair in an after school program at Roselle Park, a local middle school, using attitude surveys and concept quizzes to determine the impact of the games among the students. The pilot games were successful with the middle school students. After they had played the games, many of the girls said they could see themselves studying computer science, even though before the games, very few girls had included computer scientist as one of their two hoped for careers. Statistics gathered during the pilot indicate the need to continue this research, with more students in different demographics, and with more researcher collaboration, in order to design more adaptive games, to determine what specifically about the games influenced the girls the most, and to gain insight into how they were learning the computer concepts in the puzzles.

@article{Stewart-Gardiner:2013:IMS:2460156.2460174,
    author = {Stewart-Gardiner, Carolee and Carmichael, Gail and Latham, Jennifer and Lozano, Nathaly and Greene, Jennifer L.},
    title = {Influencing Middle School Girls to Study Computer Science Through Educational Computer Games},
    journal = {J. Comput. Sci. Coll.},
    volume = {28},
    number = {6},
    pages = {90--97},
    year = 2013,
}
                    

User control over fluid simulations is a long-standing research problem in fluids for computer graphics. Applications in games and films often require recognizable creatures or objects formed from smoke, water, or flame. This thesis describes a two-layer approach to the problem, in which a bulk flow drives a particle system towards a target distribution, while simultaneously a vortex particle simulation adds recognizable fluid motion. A bulk flow field is obtained by distributing target particles within a mesh, then matching control particles with target particles; control particles are given a trajectory bringing them to their targets, and a field is obtained by interpolating values from the control particles. A detail flow field is obtained by traditional vortex particle simulation. We render the final particle system using stochastic shadow mapping. We spend some effort optimizing our processes for speed, obtaining simulations at interactive or near-interactive rates.

@thesis{madill13smoke,
    author = {Jamie Madill},
    title = {Targeted Smoke Simulation Combining Control and Turbulent Flow},
    booktitle = {Carleton University},
    year = 2013,
}