Mosaic ATM is pleased to announce that we have been awarded a Department of Defense/Department of the Navy SBIR Phase I contract to help with maritime intelligence.
An important part of a Naval intelligence analyst’s job is to sift through a large amount of data and identify connections for decision advantage, and they spend a significant amount of time doing this.
Unfortunately, with existing tools, multimodal and streaming content is both time-consuming to analyze and cumbersome to annotate and distribute. It is largely still a manual process. For example, the large size of a video file encourages segmenting of the video data into smaller pieces containing highly valuable and sensitive information.
When this is done, however, metadata links are broken, causing the loss of temporal and geo-tracking metadata – both of which are an important part of intelligence refinement necessary to support maritime operations.
This is a problem that can be addressed using modern machine learning. SBIR N222-118, also known as Librarian: An AI Driven Multi-Int Unifying Software Tool, is our proposed solution. This advanced platform will have the ability to:
- Accept, parse, and store multimodal data sources
- Annotate the data source metadata
- Employ a collection of transformer-based neural network models that annotate and represent the content of the data as natural language or in an embedding structure that is common across all modes to allow cross-modality search
- Expose a user interface (UI) or application programming interface (API) that lets analysts easily query data across modalities.
We’re very excited about this opportunity as Librarian has significant potential for both military and civilian utility.