Earth Imagery Tomorrow
Last updated
Last updated
Technological advancements, resulting in new capabilities or a dramatic decrease in cost for the same capabilities, drive new applications and uses previously impossible. LayerDrone imagery is higher-resolution, lower-cost, and more easily updated than any other aerial imagery available today. As LayerDrone’s library of imagery expands, it may unlock new applications and uses that are still in their infancy today.
Spatial AI Model Training: High-resolution earth imagery is already being used to train geospatial AI to identify and classify different land use types, physical objects, and industry-specific assets. ESRI has incorporated geospatial AI models to assist customers with identifying large-scale land use changes; identifying infrastructure risks due to natural disasters; and predicting changes in forestation and growth. Imagery captured via the LayerDrone Network already can improve these existing models due to its higher resolution.
But even more applications of geospatial AI have yet to emerge. Large language models (LLMs) have become ubiquitous and widely used in a variety of language-based AI applications. Multi-modal LLMs have emerged as well, incorporating inputs from other modes, such as images or audio. Niantic Inc was the first to envision a large geospatial model (LGM)*, which would enable computers to understand and navigate the physical world at the same level that an LLM currently communicates using language. Creating and training an LGM will necessitate massive data inputs in the form of images from the physical world - a use case for high-resolution, consistent, widespread earth imagery. The extent of potential applications for an LGM is as-yet unknown and unimagined. In the years and decades to come, LayerDrone imagery could power the scaling up of geospatial AI to next-level spatially aware AI and beyond.
AR/VR and Metaverse Applications: With the expansion of AR/VR and real-world-based gaming, location-based geospatial data is increasingly in demand to enable creation of highly accurate digital twins. High-resolution imagery supports immersive experiences by offering accurate, real-world mapped data that enhances user experiences within both virtual and augmented environments. Consistent, up-to-date, low-cost, high resolution imagery will support the ongoing development and emergence of these experiences. The largest of these applications is the well known AR mobile gaming app PokemonGO, created by Niantic Inc, which was spun out of Google in 2015.
While these latter use cases remain in their infancy, they represent innovations that could be greatly facilitated by LayerDrone’s global high-resolution aerial imagery.
*