Cameras Viewing Cameras Geometry
This work addresses a foundational geometry problem in computer vision for applications like autonomous systems, but it appears incremental as it builds on existing multi-view geometry theory.
The paper tackles the problem of understanding scene structure from multiple images by studying the theoretical geometry of calibrated cameras that only recognize each other, addressing a timely issue due to the rise of wearable cameras, autonomous vehicles, and drones.
A basic problem in computer vision is to understand the structure of a real-world scene given several images of it. Here we study several theoretical aspects of the intra multi-view geometry of calibrated cameras when all that they can reliably recognize is each other. With the proliferation of wearable cameras, autonomous vehicles and drones, the geometry of these multiple cameras is a timely and relevant problem to study.