Autodesk Research - Recent Publications Feed

Comparison of User Performance in Mixed 2D-3D Multi-Display Environments

Tue, 26 Mar 2013 23:12:00 -0400

Abhijit Karnik, Tovi Grossman & Sriram Subramanian. (2013 - In Press).
Comparison of User Performance in Mixed 2D-3D Multi-Display Environments
INTERACT 2013 Conference Proceedings:
IFIP Conference on Human-Computer Interaction.
18 pages.

Stereoscopic displays and volumetric 3D displays capable of delivering 3D views have in use for many years. These standalone displays have been investigated in detail for their impact on users’ viewing experiences. Effects like aesthenopia and nausea are well-known for flat-screen based stereoscopic displays. However, these devices have not been tested in the context of multi-display environments (MDEs). The performance cost of repetitive switching between a 3D (stereo or volumetric) display and a standard 2D display are not known. In this paper, we perform a thorough user study where we investigate the effects of using such 3D displays within the context of a MDE. We report on our findings and discuss the implications of the same on designs involving such hybrid setups. Our experiments show that in the condition involving two 2D displays which allow for motion parallax and perspective correction, the participants performed the task the fastest.

Using General Modeling Conventions for the Shared Development of Building Performance Simulation Software

Thu, 11 Apr 2013 15:47:25 -0400

Rhys Goldstein, Simon Breslav & Azam Khan. (2013 - In Press).
Using General Modeling Conventions for the Shared Development of Building Performance Simulation Software
BS 2013 Conference Proceedings:
International Conference of the International Building Performance Simulation Association.

The building performance simulation community applies theory from several different fields to develop models for heat transfer, light propagation, human behavior, and other domains. To integrate these models, we propose the adoption of general modeling conventions from the less familiar field of modeling and simulation theory. The conventions we explore are known as the Discrete Event System Specification (DEVS). With DEVS, a model-independent simulator responsible for advancing time alleviates many of the technological difficulties involved in coupling models. We show how DEVS, from a mathematical perspective, accommodates the co-simulation strategies known as loose and strong coupling as well as strategies involving variable time steps. We also show how a model based on a functional decomposition of a system, as opposed to a topological decomposition, readily supports the sharing of domain-specific algorithms. The examples presented were implemented using Design-DEVS, an environment we created to help communities of researchers collaborate in the development of simulation software.

TutorialPlan: Automated Tutorial Generation from CAD Drawings

Mon, 15 Apr 2013 14:41:14 -0400

Wei Li, Yuanlin Zhang & George Fitzmaurice. (2013 - In Press).
TutorialPlan: Automated Tutorial Generation from CAD Drawings
IJCAI 2013 Conference Proceedings:
International Joint Conference on Artificial Intelligence.
8 pages.

Authoring tutorials for complex software applications is a time consuming process. It also highly depends on the tutorial designer’s skill level and experience. This paper introduces an approach which automatically generates software tutorials using the digital artifacts produced by the users of a software program. We model this process as an optimal planning problem using software produced artifacts, software specifications and the human-computer interaction Keystroke-Level Model (KLM). We present TutorialPlan, an automated tutorial generator, which creates step-by-step text and image instructions from CAD drawings and helps users learn AutoCAD, a complex design and drafting software. In our tutorial generator, the optimal planning problem is represented and solved using DLV, a general Answer Set Programming (ASP) system. DLV offers a natural representation of both the problem and the heuristics needed to solve it efficiently. A user study shows that the tutorials generated by our system are comparable to those generated by experienced AutoCAD users.

A Model of Navigation for Very Large Data Views

Tue, 19 Mar 2013 10:16:55 -0400

Michael Glueck, Tovi Grossman & Daniel Wigdor. (2013 - In Press).
A Model of Navigation for Very Large Data Views
GI 2013 Conference Proceedings:
Graphics Interface Conference.
8 pages.

Existing user performance models of navigation for very large documents describe trends in movement time over the entire navigation task. However, these navigation tasks are in fact a combination of many sub-tasks, the details of which are lost when aggregated. Thus, existing models do not provide insight into the navigation choices implicit in a navigation task, nor into how strategy ultimately affects user performance. Focusing on the domain of data visualizations, the very large documents we investigate are very large data views. We present an algorithmic decision process and descriptive performance model of zooming and panning navigation strategy, parameterized to account for speed-accuracy trade-offs, using common mouse-based interaction techniques. Our model is fitted and validated against empirical data, and used to evaluate proposed optimal strategies. Further, we use our model to provide support for interaction design considerations for achieving performant interaction techniques for navigation of very large data views.

Stroke Parameterization

Fri, 11 Jan 2013 15:04:00 -0500

Ryan Schmidt. (2013).
Stroke Parameterization
EUROGRAPHICS Computer Graphics Forum.
May 2013, 32(2).
8 pages.

We present a novel algorithm for generating a planar parameterization of the region surrounding a curve embedded in a 3D surface, which we call a stroke parameterization. The technique, which extends the well-known Discrete Exponential Map [SGW06], uses the same basic geometric transformations and hence is both efficient and easy-to-implement. We also handle self-intersecting curves, for which a 1-1 map between the original surface and the plane is not possible. Stroke parameterizations provide an ideal coordinate space for solving a variety of computer graphics problems. We present applications including tiling texture and displacement along 3D brush strokes, procedural texturing along 3D paths, and user-guided crease extraction.

The Effect of Time-based Cost of Error in Target-directed Pointing Tasks

Mon, 06 May 2013 13:09:07 -0400

Nikola Banovic, Tovi Grossman & George Fitzmaurice. (2013).
The Effect of Time-based Cost of Error in Target-directed Pointing Tasks
CHI 2013 Conference Proceedings:
ACM Conference on Human Factors in Computing Systems.
pp. 1373-1382.

One of the fundamental operations in today’s user interfaces is pointing to targets, such as menus, buttons, and text. Making an error when selecting those targets in real life user interfaces often results in some cost to the user. However, the existing target-directed pointing models do not consider the cost of error when predicting task completion time. In this paper, we present a model based on expected utility theory that predicts the impact of the error cost on the user’s completion time for target-directed pointing tasks. We then present a target-directed pointing user study, which results show that time-based costs of error significantly impact the user’s performance. Our results also show that users perform according to an expected completion time utility function and that optimal performance computed using our model gives good prediction of the observed task completion times.

Patina: Smart Heatmaps for Visualizing Application Usage

Mon, 06 May 2013 14:15:16 -0400

Justin Matejka, Tovi Grossman & George Fitzmaurice. (2013).
Patina: Smart Heatmaps for Visualizing Application Usage
CHI 2013 Conference Proceedings:
ACM Conference on Human Factors in Computing Systems.
pp. 3227-3236.

We present Patina, an application independent system for collecting and visualizing application usage data which requires no instrumentation of the target application for either the data collection or presentation. All data collected through standard window metrics and accessibility APIs. The primary visualization is a smart heatmap overly which is dynamically created to match the content and location of the user interface controls visible in the active application even if the controls change location or shape. After reviewing the design goals, we explain the implementation of both the data collection and presentation parts of the system. Lastly, we explore three additional application of the Patina system based on previous research and report on an internal usage evaluation.

Swifter: Improved Online Video Scrubbing

Mon, 06 May 2013 14:18:29 -0400

Justin Matejka, Tovi Grossman & George Fitzmaurice. (2013).
Swifter: Improved Online Video Scrubbing
CHI 2013 Conference Proceedings:
ACM Conference on Human Factors in Computing Systems.
pp. 1159-1168.

Online streaming video systems have become extremely popular, yet navigating to target scenes of interest can be a challenge. While recent techniques have been introduced to enable real-time seeking, they break down for large videos, where scrubbing the timeline causes video frames to skip and flash too quickly to be comprehendible. We present Swifter, a new video scrubbing technique that displays a grid of pre-cached thumbnails during scrubbing actions. In a series of studies, we first investigate possible design variations of the Swifter technique, and the impact of those variations on its performance. Guided by these results we compare an implementation of Swifter to the previously published Swift technique, in addition to the approaches utilized by YouTube and Netfilx. Our study finds that Swifter significantly outperforms each of these techniques in a scene locating task, by a factor of up to 48%.

A Multi-Site Field Study of Crowdsourced Contextual Help: Usage and Perspectives of End-Users and Software Teams

Mon, 06 May 2013 12:44:48 -0400

Parmit Chilana, Andrew Ko, Jacob Wobbrock & Tovi Grossman. (2013).
A Multi-Site Field Study of Crowdsourced Contextual Help: Usage and Perspectives of End-Users and Software Teams
CHI 2013 Conference Proceedings:
ACM Conference on Human Factors in Computing Systems.
pp. 217-226.

We present a multi-site field study to evaluate LemonAid, a crowdsourced contextual help approach that allows users to retrieve relevant questions and answers by making selections within the interface. We deployed LemonAid on 4 different web sites used by thousands of users and collected data over several weeks, gathering over 1,200 usage logs, 168 exit surveys, and 36 one-on-one interviews. Our results indicate that over 70% of users found LemonAid to be helpful, intuitive, and desirable for reuse. Software teams found LemonAid easy to integrate with their sites and found the analytics data aggregated by LemonAid a novel way of learning about users’ popular questions. Our work provides the first holistic picture of the adoption and use of a crowdsourced contextual help system and offers several insights into the social and organizational dimensions of implementing such help systems for real-world applications.

Community Enhanced Tutorials: Improving Tutorials with Multiple Demonstrations

Mon, 06 May 2013 14:20:40 -0400

Ben Lafreniere, Tovi Grossman & George Fitzmaurice. (2013).
Community Enhanced Tutorials: Improving Tutorials with Multiple Demonstrations
CHI 2013 Conference Proceedings:
ACM Conference on Human Factors in Computing Systems.
10 pages.

Web-based tutorials are a popular help resource for learning how to perform unfamiliar tasks in complex software applications. However, in their current form, web tutorials are isolated from the applications that they support. In this paper, we present FollowUs, a new web-tutorial system that integrates a full-featured application into a web-based tutorial. This novel architecture enables community enhanced tutorials, which continuously improve as more users work with them. FollowUs captures video demonstrations of users as they perform a tutorial. Subsequent users can then view the author’s tutorial, or choose from a library of captured community demonstrations for each step of the tutorial. A user study showed that user’s perform significantly better when using our system with a library of multiple demonstrations, in comparison to its equivalent baseline system with only the original authored content.