An Application-Agnostic Automatic Target Recognition System Using Vision Language Models

TL;DR

This paper introduces a flexible, open-vocabulary ATR system leveraging vision-language models, allowing non-technical users to define target classes at runtime using natural language or images, with enhanced performance techniques.

Contribution

The novel ATR system enables target definition at runtime through natural language or images and integrates multiple techniques for improved detection and visualization.

Findings

Effective detection of targets with minimal training data

Enhanced performance through tubelet identification and linking

Visualizations include mosaics and heatmaps of detected areas

Abstract

We present a novel Automatic Target Recognition (ATR) system using open-vocabulary object detection and classification models. A primary advantage of this approach is that target classes can be defined just before runtime by a non-technical end user, using either a few natural language text descriptions of the target, or a few image exemplars, or both. Nuances in the desired targets can be expressed in natural language, which is useful for unique targets with little or no training data. We also implemented a novel combination of several techniques to improve performance, such as leveraging the additional information in the sequence of overlapping frames to perform tubelet identification (i.e., sequential bounding box matching), bounding box re-scoring, and tubelet linking. Additionally, we developed a technique to visualize the aggregate output of many overlapping frames as a mosaic of the area scanned during the aerial surveillance or reconnaissance, and a kernel density estimate (or heatmap) of the detected targets. We initially applied this ATR system to the use case of detecting and clearing unexploded ordinance on airfield runways and we are currently extending our research to other real-world applications.