Contextual Interaction via Primitive-based Adversarial Training For Compositional Zero-shot Learning

TL;DR

This paper introduces a primitive-based adversarial training approach for compositional zero-shot learning, effectively modeling primitive interactions and improving performance on standard datasets.

Contribution

It proposes a novel, model-agnostic adversarial training method and an over-sampling strategy with object-similarity guidance for better primitive interaction modeling in CZSL.

Findings

Achieved state-of-the-art results on UT-Zappos50K, MIT-States, and C-GQA datasets.

Demonstrated improved recognition of novel attribute-object compositions.

Validated effectiveness through detailed quantitative and retrieval experiments.

Abstract

Compositional Zero-shot Learning (CZSL) aims to identify novel compositions via known attribute-object pairs. The primary challenge in CZSL tasks lies in the significant discrepancies introduced by the complex interaction between the visual primitives of attribute and object, consequently decreasing the classification performance towards novel compositions. Previous remarkable works primarily addressed this issue by focusing on disentangling strategy or utilizing object-based conditional probabilities to constrain the selection space of attributes. Unfortunately, few studies have explored the problem from the perspective of modeling the mechanism of visual primitive interactions. Inspired by the success of vanilla adversarial learning in Cross-Domain Few-Shot Learning, we take a step further and devise a model-agnostic and Primitive-Based Adversarial training (PBadv) method to deal with this problem. Besides, the latest studies highlight the weakness of the perception of hard compositions even under data-balanced conditions. To this end, we propose a novel over-sampling strategy with object-similarity guidance to augment target compositional training data. We performed detailed quantitative analysis and retrieval experiments on well-established datasets, such as UT-Zappos50K, MIT-States, and C-GQA, to validate the effectiveness of our proposed method, and the state-of-the-art (SOTA) performance demonstrates the superiority of our approach. The code is available at https://github.com/lisuyi/PBadv_czsl.