IARPA SMART Public Dataset

The IARPA Space-Based Machine Automated Recognition Technique (SMART) program was one of the first large-scale research program to advance the state of the art for automatically detecting, characterizing, and monitoring large-scale anthropogenic activity in global scale, multi-source, heterogeneous satellite imagery. The program leveraged and advanced the latest techniques in artificial intelligence (AI), computer vision (CV), and machine learning (ML) applied to geospatial applications. The Johns Hopkins Applied Physics Lab (JHU/APL) led the development of a large, global-scale dataset containing spatio-temporal annotations of large scale heavy construction activity for the purposes of algorithm development and evaluation for automated broad area search and classification of anthropogenic activities from satellite imagery. This dataset contains these spatio-temporal annotations. 

A semi-custom implementation of performance evaluation metrics tailored specifically to the data formats and application introduced by this dataset is also available on our Github page.