Problem Statement

The objective of the proposed project is to develop vision language models that provide hierarchical keywords from the imagery data and approaches to integrate these keywords with other reasoning features in the imagery data for explainable autonomous systems. 

Technical Approach: 

• Identify a specific autonomous vision task.
• Establish the relevant language data domain for the research. 
• Develop hierarchical keyword vision language model with appropriate language structure and format 
• Formulate strategies to select and present reasoning and interpretation in both language format and other effective means from the integrated model.