Graph based Dynamic Segmentation of Generic Objects in 3D
This addresses the challenge of segmenting generic objects in 3D for applications like robotics or augmented reality, but appears incremental as it builds on existing 3D segmentation approaches.
The paper tackles the problem of 3D object segmentation in generic scenarios with frequent object interactions using RGB-D stream data, proposing a novel tree structure representation and a dynamic management mechanism for connected components, resulting in a method that is generic and does not require initialization.
We propose a novel 3D segmentation method for RBGD stream data to deal with 3D object segmentation task in a generic scenario with frequent object interactions. It mainly contributes in two aspects, while being generic and not requiring initialization: firstly, a novel tree structure representation for the point cloud of the scene is proposed. Then, a dynamic manangement mechanism for connected component splits and merges exploits the tree structure representation.