Global Structural Knowledge Distillation for Semantic Segmentation

Knowledge distillation (KD) has become a cornerstone for compressing deep neural networks, allowing a smaller student model to learn from a larger teacher model. In the context of semantic segmentation, traditional KD methods primarily focus on pixel-level feature alignment, where the student model...

全面介绍

书目详细资料

Principais autores:	Hyejin Park, Keonhee Ahn, Hyesong Choi, Dongbo Min
格式:	Artigo
语言:	Inglês
出版:	IEEE 2025-01-01
丛编:	IEEE Access
主题:	Knowledge distillation model compression semantic segmentation image classification global structural similarity class-balanced sampling
在线阅读:	https://ieeexplore.ieee.org/document/11018413/
标签:	添加标签 没有标签, 成为第一个标记此记录!