"Semantic 3D Object Maps for Everyday Manipulation in Human Living Environments" is a Ph. D. thesis by Radu B. Rusu, in visual perception for robots. This thesis proposes Semantic 3D Object Models as a novel representation of the robot's operating environment that suffice the needs of autonomous robotic platforms, and shows how the relevant information can be automatically acquired from dense 3D range data.

The cover I made is based on the material I read about the algorithms and procedures described in the book itself. The image is an artistical rendition of how the robot perceives the environment, and it also shows the different computational phases of the environment analysis: point cloud data acquisition, optimisation algorithms, objects detection, conversion of data into a mesh (wireframe) and the end result which is used for the task at hand. The color palette I chose is meant to be daring, something different, and to remind the viewers of SciFi movies.

Download this 3D robot and other 3 robot models (ZIP file, 4.5 mb)