Visual search over billions of aerial and satellite images
This enables real-time visual search over the Earth's surface, which is useful for applications in geospatial analysis and remote sensing, though it is incremental as it applies existing methods to a new large-scale dataset.
The paper tackles the problem of performing visual search over billions of aerial and satellite images by using a convolutional neural network to define visual similarity and a hash-based search system, achieving a search time of approximately 0.1 seconds for 2 billion images.
We present a system for performing visual search over billions of aerial and satellite images. The purpose of visual search is to find images that are visually similar to a query image. We define visual similarity using 512 abstract visual features generated by a convolutional neural network that has been trained on aerial and satellite imagery. The features are converted to binary values to reduce data and compute requirements. We employ a hash-based search using Bigtable, a scalable database service from Google Cloud. Searching the continental United States at 1-meter pixel resolution, corresponding to approximately 2 billion images, takes approximately 0.1 seconds. This system enables real-time visual search over the surface of the earth, and an interactive demo is available at https://search.descarteslabs.com.