The motivation of the project is to apply computer vision techniques in underground mining conditions to improve the performance of machines and reduce human dependency in a dangerous environment. Following are the objectives we wish to solve in my capstone project.
Multi-camera Pose estimation
Given a set of cameras and their known intrinsic parameters, estimate the 6DoF pose ( i.e., Calculate Rotation and Translation parameters) of each of the cameras present in the setting.
Given a set of cameras placed at a distance create a composite panoramic image by blending/ stitching the images captured from the cameras.
3D depth estimation
Given a set of cameras, estimate the 3D depth of the scene.
Though these topics are fairly explored in the computer vision domain, an important challenge in this project is to implement these techniques in underground coal seam mining conditions. The images obtained are of a mostly flat surface with low illumination.