| Nominations | Video Details with Developer URLs | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Best Video | Best Educational Video | Best Short Video | Best Student Video | Best Demonstration Video | Most Innovative Video | Title | Developer(s) | Affiliation(s) | Description | Length | URL (Size MB) |
| x | Westworld: A Domain for Automatically Generated Narratives | Mike Dominguez, Arnav Jhala | North Carolina State University | Darshak is a system developed at NCSU that features a narrative discourse planning algorithm for generating cinematic visualizations of automatically generated stories. The characters and cameras are directed through a suite of tools built on the Unreal Tournament game engine. Darshak automatically creates movies in WestWorld, a western themed environment in the game engine, featuring content from the LawDogs Unreal mod. | 0:56 |
MOV (24) |
|||||
| x | x | Digital Analysis of Van Gogh Paintings | Laurens van der Maaten, Eric Postma | Maastricht University | Computer processing of digital images of artwork is an emerging and rapidly growing cross-disciplinary activity. The video presents a prototype of a system that can assist art experts in the authenticity assessment of alleged Van Gogh paintings. The presented system is based on a combination of texton-based texture analysis and t-Distributed Stochastic Neighbor Embedding. (Winner, Most Innovative Video) | 4:50 |
MP4 (56) |
||||
| x | Bio-inspired Flying Robots | Sabine Hauert | Laboratory of Intelligent Systems, EPFL | In this video we present bio-inspiration as the main drive for the design of platforms, autonomous navigation strategies and aerial swarming for flying robots. (Winner, Best Video) | 4:59 |
MOV (150) |
|||||
| The Secure Wireless Agent Testbed | Evan Sultanik, Michael Sausa, Arnol Sathe, Kristy Pron, Seth Kendall, with William Regli as Faculty Advisor | Drexel University | This video presents a high-level overview of the Secure Wireless Agent Testbed, a multi-year project to study intelligent agent technologies in the context of network-centric and mobile ad hoc wireless systems for situation awareness. The project involved over 50 students and resulted in over two dozen live field trials of agent-based C4ISR. The SWAT platform has since been employed on a number of follow-on projects and products for the Department of Defense, Department of Justice and several commercial offerings. | 5:00 |
MOV (208) Streaming |
||||||
| x | BigDog | Marc Raibert, Kevin Blankespoor, Gabe Nelson, Rob Playter and the Entire BigDog Team | Boston Dynamics | The video shows recent testing of BigDog, a rough-terrain quadruped robot developed by Boston Dynamics with funding from DARPA. The goal for BigDog is to carry equipment on any terrain where humans can go on foot. (BigDog was developed by Boston Dynamics with funding from DARPA.) (Winner, Best Short Video) | 1:00 |
WMV (23) |
|||||
| CALO Integrated Task Learning | Will Haines, Aaron Spaulding | SRI, USC ISI | The CALO Integrated Task Learning (ITL) video describes an effort to develop technology that can learn and execute procedures to automate online tasks in a range of application domains. In this video, the technology developers discuss both their approach to task learning, which is grounded in the notion of a user demonstrating and describing the task to the system, and challenges faced in deploying the technology. A short demonstration shows a user teaching the ITL system how to find driving directions to SRI's main campus and email them to interested parties. | 4:59 |
WMV (10) |
||||||
| x | CALO Workflow Recognition and Proactive Assistance | H. H. Bui, F. Cesari, D. Elenius, N. House, D. N. Morley, K. M. Myers, S. Natarajan, S. Saadati, E. Yeh, N. Yorke-Smith | SRI | SRI's CALO agent helps users manage their tasks and time. Using advanced activity recognition algorithms based on logical probabilistic models, CALO observes and tracks user desktop actions and offers contextually appropriate assistance. | 1:00 |
AVI (91) |
|||||
| DARPA LAGR Program: Learning Applied to Long-Range Vision using a Collision-Free Navigation Platform | Pierre Sermanet, Raia Hadsell, Marco Scoffier, Matt Grimes, Jan Ben, Ayse Erkan, Chris Crudele, Urs Muller, Yann LeCun | New York University | New York University uses machine learning to extend the vision of the DARPA LAGR robots. | 5:00 |
AVI (182) |
||||||
| Automatically Generated Convex Regions for Agent Navigation in Virtual Worlds | D. Hunter Hale, G. Michael Youngblood | University of North Carolina, Charlotte | We demonstrate a new method of decomposing virtual environments into a navigation mesh represented by bounding geometry useful for agent spatial reasoning, path planning, and mobility. Our method is a fully-automated and tunable growth-based technique for decomposing the world using quads and higher order polygonal shapes. This technique results in decompositions that are generally better than current techniques. | 5:00 |
MOV (252) |
||||||
| x | Understanding Agent Behaviors in Game Environments | Priyesh N. Dixit, G. Michael Youngblood | University of North Carolina, Charlotte | Visual representation of logged game player data from both real and artificial agents can provide insight in order to discover patterns and behavior that would otherwise be hidden. We explore the use of visual data mining and automated data processing to investigate the logged player data and present the results of finding five unique phenomena in our dataset of 3079 players. These phenomena include pirouettes, flusters, jumpers, learning, and emergent behavior. Using the visual data mining process, we can find occurrences of such behavioral patterns in order to understand player behavior and improve the interactive experience. | 4:39 |
MOV (159) |
|||||
| x | Tele-Immersive Dance: A Human/Computer Creative Environment | Renata Sheppard, Klara Nahrstedt, Ruzena Bajcsy, Lisa Wymore | University of Illinois, Urbana-Champaign | Tele-immersive Dance (TED) is a computer science and dance collaboration focused on building a computational model for human creativity. Within the 3D virtual room of TED, a dancer from Illinois is able to interact, fully immersed with a dancer from California using multiple points of view. These cross-continental collaborations reveal the potential of TED as a creativity tool for human/computer interaction. The next phase of research involves building a computational model of human creativity that will translate visual data signals into symbols that can then generate a computer response for the user. | 3:41 |
HTML for MOV (297) |
|||||
| x | CB2: Child Robot with Biomimetic Body | Tomoyuki Noda, Shuhei Ikemoto, Daniel Quevedo, Toshihiko Shimizu, Hidenobu Sumioka, Hisashi Ishihara, Yuki Sasamoto, Yuichiro Yoshikawa, Takashi Minato, Hiroshi Ishiguro, Minoru Asada | JST ERATO, Osaka University, Spray Studios | CB2, a novel child-like robot, has been built by the JST ERATO Asada Project to simulate child development. Its flexible, life-like movement and full-body soft skin enable it to interact closely with people. This video demonstrates the robot's potential as a synthetic platform for understanding the development of human communication. | 5:00 | MPG (47) | |||||
| x | x | Game-Based Learning | Daniel Byers, Michael Lapping-Carr, Julie Kumar, Thea Hinkle, Dan Grollman, Chad Jenkins | Brown University | On a mission to teach their Aibo robo-dogs to play soccer, the Brown Robotics Group is designing an online video game experience where users can remotely control the dogs in real soccer matches. The data logged from the games will be used to teach these old dogs some brand new tricks! (Winner, Best Student Video) | 4:30 |
MOV (358) |
||||
| Hands-Free Human-Robot Interaction | Nathan Koenig, Matthew Loper, Sonia Chernova, Chris Jones, Chad Jenkins, Daniel Byers | Brown University, iRobot | Using emerging sensing technology, the iRobot Packbot is demonstrated responding to voice and gesture-based commands in real time. | 3:43 |
AVI (39) |
||||||
| Konabot | Diego Ariel Bendersky, Andrea Roxana Katz, Juan Miguel Santos, Angel Juan Siri, Sergio Anibal Soria, Andres Gaston Stoliar, Paula Bassi, Armando Doria | University of Buenos Aires | Konabot is the first Explosives Robot created entirely in Argentina. The project was developed by researchers of the Department of Computation (Facultad de Ciencias Exactas y Naturales de la Universidad de Buenos Aires). In this video the researchers explain the project development and how Konabot works. | 5:00 |
MPG (182) |
||||||
| x | The Sounds of Speech | John Rubin | John Rubin Productions | The Sounds of Speech (Rubin, 2002) is a segment excerpted from the Reading Rocket series commissioned by the United States Department of Education. It shows the Reading Tutor developed by Carnegie Mellon University's Project LISTEN (www.cs.cmu.edu/~listen), directed by Jack Mostow. The Reading Tutor uses automatic speech recognition to listen to children read aloud, and responds with recorded human speech and graphical feedback modeled after reading experts but adapted to the affordances and limitations of speech technology. | 3:20 |
MPG (33) Streaming |
|||||
| x | Narratoria, an Authoring Suite for Digital Interactive Narrative | Martin van Velsen | USC Institute for Creative Technologies | The software presented here provides sufficient authoring capabilities to allow users to create training applications using interactive narrative without the need of programmers. Novices and professionals alike can use the system's toolset to manage and create complex interactive storylines. | 3:15 | AVI (192) | |||||
| Two Generations of Robots for Human-Robot Interaction Research | Candace L. Sidner, Christopher Lee, Charles Rich | BAE Systems AIT, Boston Dynamics, Worcester Polytechnic Institute | In our research on engagement and collaboration in human-robot interaction at Mitsubishi Electric Research Laboratories (MERL) from 2003 to 2007, we developed first a "penguinoid" and then a humanoid robot which can recognize and track human faces, detect and respond to human nods and gaze, and which itself uses pointing, gaze and simple spoken dialogue to communicate. This research is continuing at WPI. | 4:20 |
MOV (88) |
||||||
| Smart Classroom | Saurabh Goyal, Chaitanya Mutyala, Purnima Mandava, Srinagadeepthi Paladugu | University of Missouri, Kansas City | Smart Classroom is an exciting and visionary technology designed to improve the quality of teaching and learning experiences at colleges and universities. This video has been created as part of the course work for "Creative Design for Software Systems" at the University of Missouri - Kansas City. | 4:40 |
MPG (193) |
||||||
| x | x | Virtual Humans Research Project | Patrick Kenny, Jonathan Gratch, Bill Swartout | USC Institute for Creative Technologies | This video features virtual human research at the USC Institute for Creative Technologies. The virtual humans use artificial intelligence to reason about events as they unfold, they exhibit emotions, and they interact naturally using verbal and non-verbal communication. This video features a prototype system that shows how virtual humans can be used to train complex interpersonal skills, such as negotiation | 3:04 |
WMV (37MB) MOV (88MB) |
||||
| x | x | Imitating a RoboCup Soccer Player Using Case-Based Reasoning | Michael Floyd, Babak Esfandiari | Carleton University | This video shows how case-based reasoning can be used to imitate a RoboCup soccer agent. We describe how the imitating agent learns through observation and how it imitates. Also, the video shows the behaviour of the imitating agent at various stages in the learning process. (Winner, Best Educational Video) | 5:00 |
MP4 (28) |
||||
| The Tactical Language and Culture Training System | Andre Valente, W. Lewis Johnson, James Reilly | Alelo, Inc. | This video shows the Tactical Language and Culture training systems and briefly presents some of the ways they use AI technology. The systems are an example of successful technology transition from academia to practice. They have been deployed widely in the US Department of Defense, and more recently in commercial applications. | 4:45 | WMV (24) | ||||||
| x | Kato, the Geocaching Robot | Michael Sokolsky, James Neufeld, Michael Bowling | University of Alberta | Kato is a geocaching robot at the University of Alberta. This video shows how Kato sees the world with its laser scanner and how it uses this information to navigate around obstacles to a hidden cache. | 4:35 |
MOV (214) |
|||||
| x | Social Referencing with Leonardo | Matt Berlin, Jesse Gray, Andrea Thomaz, Cynthia Breazeal | MIT Media Lab | The furry robot Leonardo engages in an interaction with a human teacher based on infant social referencing. Leo learns how to react to a number of novel objects by attending to the emotional reactions of his caregiver | 2:44 |
MOV (62) |
|||||
| x | CBR Noir | David Wilson | University of North Carolina, Charlotte | CBR Noir is a high-level, tongue-in-cheek introduction to Case-Based Reasoning (CBR). It is presented in the style of a classic Film Noir murder mystery. For the connoisseur, the dialogue incorporates the names of many early CBR systems; see if you can count them all before the credits roll. (Winner, Best Video Narration) | 3:51 |
MPG (47) |
|||||
| x | x | STARMAC: Stanford Testbed of Autonomous Rotorcraft for Multi-Agent Control | Gabriel Hoffmann, Steven Waslander, Haomiao Huang, Vijay Pradeep, Michael Vitus, Jeremy Gillula, Claire Tomlin | Stanford University | STARMAC, the Stanford Testbed of Autonomous Rotorcraft for Multi-Agent is a testbed of 6 quadrotor helicopter unmanned aerial vehicles that fly indoors and outdoors to experiment with autonomous agent algorithms. This video presents an overview of the control system, the vehicle capabilities, and the experimental applications for the testbed. Results of many experimental flights are shown, along with some fun flight results that push the limits of what the vehicles should be able to handle. (Winner, Best Demonstration Video) | 4:48 |
WMV (29) |
||||