Current methods for Earth observation tasks such as semantic mapping, map alignment, and change detection rely on near-nadir images; however, often the first available images in response to dynamic world events such as natural disasters are oblique.
In this work, we develop machine learning models that use satellite imagery to perform indirect top-down estimation of road transport emissions.
An object's geocentric pose, defined as the height above ground and orientation with respect to gravity, is a powerful representation of real-world structure for object detection, segmentation, and localization tasks using RGBD images.
We introduce a deep learning approach to perform fine-grained population estimation for displacement camps using high-resolution overhead imagery.
The increasingly common use of incidental satellite images for stereo reconstruction versus rigidly tasked binocular or trinocular coincident collection is helping to enable timely global-scale 3D mapping; however, reliable stereo correspondence from multi-date image pairs remains very challenging due to seasonal appearance differences and scene change.
Autonomously searching for hazardous radiation sources requires the ability of the aerial and ground systems to understand the scene they are scouting.
We introduce the novel problem of determining the relevance of questions to images in VQA.
Our approach produces a diverse set of plausible hypotheses for both semantic segmentation and prepositional phrase attachment resolution that are then jointly reranked to select the most consistent pair.