This Project Grant award, valued at $111,878, was provided by the National Science Foundation under the Computer and Information Science and Engineering (CFDA 47.070) program. The award aims to develop a hybrid, vision-centric framework that integrates intuitive and deliberate visual processing methods to create more robust visual intelligence. The key objectives are to:
-
Advance vision-centric parametric knowledge to form the foundational layer of intuitive understanding through techniques like visual self-supervised learning, language guidance, and generative modeling.
-
Incorporate human-like non-parametric mechanisms, such as visual search and working memory, to enhance deliberate reasoning capabilities.
-
Integrate the parametric and non-parametric approaches into a unified, hybrid architecture that can activate either method as needed for task-specific demands.
The framework will be tested in real-world, dynamic environments beyond static image datasets, including tasks requiring long-form video analysis and visual-spatial reasoning. This research aims to produce more adaptable, reliable, and broadly useful vision-based applications. The award was granted to New York University, a leading research institution, and has an ultimate completion date of January 31, 2030.
Generated 4/1/25, 4:58 AM