Joint graph entropy knowledge distillation for point cloud classification and robustness against corruptions
Abstract
Classification tasks in 3D point clouds often assume that class events are follow independent and identically distributed (IID), although this assumption destroys the correlation between classes. This study paper proposes a classification strategy, Joint Graph Entropy Knowledge Distillation (JGEKD), suitable for non-independent and identically distributed 3D point cloud data, which the strategy achieves knowledge transfer of class correlations through knowledge distillation by constructing a loss function based on joint graph entropy. Firstly, we employ joint graphs to capture addthe hidden relationships between classes and, implement knowledge distillation to train our model by calculating the entropy of addadd graph. Subsequently Then, to handle 3D point clouds that is invariant to spatial transformations, we construct Ssiamese structures and develop two frameworks, self-knowledge distillation and teacher-knowledge distillation, to facilitate information transfer between different transformation forms of the same data. In addition Additionally, we use the above framework to achieve knowledge transfer between point clouds and their corrupted forms, and increase the robustness against corruption of model. Extensive experiments on ScanObject, ModelNet40, ScanntV2\cls and ModelNet-C demonstrate that the proposed strategy can achieve competitive results.
Turn this paper into a full lesson
ArcXiv compiles a staged curriculum from this paper: 8-12 lessons across beginner → advanced, synthesised section guides, visuals, flashcards, a quiz, exercises, and on-demand deep dives per section. Grounded in the abstract, never invented.