Mutually Causal Semantic Distillation Network for Zero-Shot Learning

TL;DR

This paper introduces MSDN++, a novel mutual causal attention network that enhances zero-shot learning by effectively capturing intrinsic semantic relationships between visual and attribute features, leading to state-of-the-art results.

Contribution

The paper proposes a mutually causal semantic distillation network with bidirectional causal attention mechanisms for improved semantic knowledge transfer in ZSL.

Findings

Achieves new state-of-the-art performance on benchmark datasets.

Significantly outperforms previous methods in zero-shot classification accuracy.

Demonstrates the effectiveness of mutual causal attention in semantic feature learning.

Abstract

Zero-shot learning (ZSL) aims to recognize the unseen classes in the open-world guided by the side-information (e.g., attributes). Its key task is how to infer the latent semantic knowledge between visual and attribute features on seen classes, and thus conducting a desirable semantic knowledge transfer from seen classes to unseen ones. Prior works simply utilize unidirectional attention within a weakly-supervised manner to learn the spurious and limited latent semantic representations, which fail to effectively discover the intrinsic semantic knowledge (e.g., attribute semantic) between visual and attribute features. To solve the above challenges, we propose a mutually causal semantic distillation network (termed MSDN++) to distill the intrinsic and sufficient semantic representations for ZSL. MSDN++ consists of an attribute$\rightarrow$visual causal attention sub-net that learns attribute-based visual features, and a visual$\rightarrow$attribute causal attention sub-net that learns visual-based attribute features. The causal attentions encourages the two sub-nets to learn causal vision-attribute associations for representing reliable features with causal visual/attribute learning. With the guidance of semantic distillation loss, the two mutual attention sub-nets learn collaboratively and teach each other throughout the training process. Extensive experiments on three widely-used benchmark datasets (e.g., CUB, SUN, AWA2, and FLO) show that our MSDN++ yields significant improvements over the strong baselines, leading to new state-of-the-art performances.