The purpose of the Applied Imagery Pattern Recognition (AIPR) annual workshops is to bring together researchers from government, industry, and academia in an elegant setting conducive to technical interchange across a broad range of disciplines. The papers span a range of topics, from research to fielded systems and provide to scientists, developers, and managers alike, a broad vision of the applicability of image analysis and machine learning technologies.
Human interactions with artificial intelligence agents is undergoing unprecedented revolutionary advances that will transform the future of society. Global social media networks, nearly unlimited bandwidth and sharing of imagery, structured multimedia, and unstructured knowledge have spurred collaboration in unexpected ways. Mobility and cloud services have enabled rapid collection and dissemination of high volume, imaging and multimodal sensor data integrated with low latency database services with increasing AI sophistication. Advancing the state-of-the-art in AI learning in complex data environments, deploying AI systems in novel and unfamiliar environments will be critical for developing scalable cognitive systems. Resource constrained AI accessible at the point of need with more generalizable, explainable, and predictable AI for imaging and video will enhance the utility of cognitive and collaborative systems.
Petabytes of imagery and multimodal datasets describing environmental, cultural, civil, and political institutions are now publically available on a global scale. Cloud resources have enabled the integration of diverse datasets and collaboration among individuals across the globe and spanning the gamut of scientific and engineering disciplines. The power of collaboration both human-human and human-machine enables integration across cultural experiences and offers the promise for expanding cultural identities and artifacts. Cognitive and computational algorithms are evolving and adapt to different learning styles and backgrounds, preserve and share knowledge globally across linguistic boundaries. Current cognitive studies are realizing human activities and responses to enable more realistic simulations. The 2019 IEEE AIPR Workshop will explore these cognitive applications of vision, dynamic scene understanding, machine learning, the associated supporting applications, and the system engineering to support the dynamic workflows.
The Workshop Committee invites papers that address any aspects of how computational cognition has used human collaboration to improve the science of pattern recognition, development of novel tools, and theory and mechanics of computational cognition. Topics include, but are not limited to, the following:
- Multi-agent systems, Cognitive & Explanatory AI
- Cooperative Human-Robot Intelligence, Transfer Learning for Skill Acquisition
- Autonomous Driving Systems
- Organically Adaptive Deep Learning for Novel Environments and Situations
- Human-Machine Collaborative Exploration of Dynamic Environments
- Remote Sensing and Autonomy
- Collaborative Learning Between Human and Robotic Systems
- Visual Cloud Computing
- Block Chains and Imaging to Track Computational Cognition and Provenance
- Medical Applications, Hyperspectral Data Fusion
- Building of computational cognition from simple learning strategies
- Transfer learning of behaviors
- And of course, advancing pattern recognition and image analysis
Keynote speakers include:
- Lieutenant General John N.T. “Jack” Shanahan, USAF, Director of the Joint Artificial Intelligence Center (JAIC)
- Dr. Cindy Daniell, Director of Research for the National Geospatial-Intelligence Agency
- Col Mike Fazen, Former Army Executive Consultant to the Director of DARPA
- Dr. Joseph Evans, Director of the Geospatial Cloud Anlaytic (GCA) Program at DARPA
Banquet Speaker: David L. Olin, “The conservation of certain iconic paintings and murals - Engineering, analysis and an objective interpretation”
TUTORIAL: In addition, there will be a half-day Tutorial: Visual Deep Learning: Current Status and Future Outlook
Description: This hands-on tutorial will describe the current capabilities of Deep Learning networks for object detection and recognition, including some practical rules of thumb for understanding when Deep Learning is and isn't (yet) likely to work. Led by David Crandall and Robert Pless, the tutorial will give a brief background on modern Deep Learning tools and approaches, approaches to visualize what Deep Learning architectures have learned from their training data, and case studies of what was necessary to build and deploy large scale practical systems that address large scale practical challenges.
Deadline for abstracts: 1 July 2019. The Workshop will include oral and poster presentations, several keynote talks that provide in-depth overviews of the fields, and a debate between two experts on a fundamental question. Written papers will be required (due after the workshop) and will be indexed in IEEE Xplore. AIPR 2019, the 48th annual workshop, is sponsored by the IEEE Computer Society Technical Committee on Pattern Analysis and Machine Intelligence, and organized by the AIPR Workshop Committee with generous support from other sponsors.
Best Student Paper and Poster Presentation Awards
AIPR 2019 will recognize and award two prizes - one for the best student paper presentation, and one for the best student poster presentation, as judged by a special awards committee. Each winner will be awarded a prize of $100. A paper or poster is eligible if (1) the first author is a student at the time of paper submission, and (2) this person will indeed present the paper or poster at the workshop if it is accepted. A student may submit in each of the categories, but can win an award in at most one of them.