Stacked Semantic-Guided Attention Model for Fine-Grained Zero-Shot Learning

TL;DR

This paper introduces a stacked semantics-guided attention model that enhances fine-grained zero-shot learning by focusing on discriminative local regions guided by class semantics, leading to improved classification and retrieval performance.

Contribution

The paper proposes a novel stacked semantics-guided attention model that uses class semantic features to guide visual attention for better fine-grained ZSL performance.

Findings

Consistent improvement on CUB and NABird datasets

Enhanced fine-grained zero-shot classification accuracy

Improved retrieval results in experiments

Abstract

Zero-Shot Learning (ZSL) is achieved via aligning the semantic relationships between the global image feature vector and the corresponding class semantic descriptions. However, using the global features to represent fine-grained images may lead to sub-optimal results since they neglect the discriminative differences of local regions. Besides, different regions contain distinct discriminative information. The important regions should contribute more to the prediction. To this end, we propose a novel stacked semantics-guided attention (S2GA) model to obtain semantic relevant features by using individual class semantic features to progressively guide the visual features to generate an attention map for weighting the importance of different local regions. Feeding both the integrated visual features and the class semantic features into a multi-class classification architecture, the proposed framework can be trained end-to-end. Extensive experimental results on CUB and NABird datasets show that the proposed approach has a consistent improvement on both fine-grained zero-shot classification and retrieval tasks.