Class-relation Knowledge Distillation for Novel Class Discovery

2 code implementations • ICCV 2023 • Peiyan Gu, Chuyu Zhang, Ruijie Xu, Xuming He

In addition, to enable a flexible knowledge distillation scheme for each data point in novel classes, we develop a learnable weighting function for the regularization, which adaptively promotes knowledge transfer based on the semantic similarity between the novel and known classes.

Knowledge Distillation Novel Class Discovery +4