Graph-based semi-supervised learning with Local Binary Patterns for holistic object categorization
作者：
Highlights：
• We propose a graph construction scheme that is based on data self-representation.
• The approach adaptively provides graphs without parameter tuning.
• The proposed scheme uses locality-constrained and sparsity encouraged coding.
• Semi-supervised learning experiments were conducted on indoor and outdoor scenes.
• The proposed method can outperform competing methods.
摘要
•We propose a graph construction scheme that is based on data self-representation.•The approach adaptively provides graphs without parameter tuning.•The proposed scheme uses locality-constrained and sparsity encouraged coding.•Semi-supervised learning experiments were conducted on indoor and outdoor scenes.•The proposed method can outperform competing methods.
论文关键词：Graph-based semi-supervised learning,Graph-based label propagation,Local Binary Patterns,Holistic object classification,Outdoor scenes,Indoor scenes
论文评审过程：Available online 22 June 2014.
论文官网地址：https://doi.org/10.1016/j.eswa.2014.06.025
