CVAug 19, 2024

Graph-guided Cross-composition Feature Disentanglement for Compositional Zero-shot Learning

arXiv:2408.09786v21 citationsh-index: 11Has Code
AI Analysis

This work addresses a specific bottleneck in CZSL for AI systems, offering an incremental improvement over existing methods.

The paper tackles the challenge of learning generalizable primitive features across different compositions in Compositional Zero-shot Learning by proposing cross-composition feature disentanglement, which leverages a compositional graph and adapters on CLIP, resulting in significant performance improvements on three benchmarks.

Disentanglement of visual features of primitives (i.e., attributes and objects) has shown exceptional results in Compositional Zero-shot Learning (CZSL). However, due to the feature divergence of an attribute (resp. object) when combined with different objects (resp. attributes), it is challenging to learn disentangled primitive features that are general across different compositions. To this end, we propose the solution of cross-composition feature disentanglement, which takes multiple primitive-sharing compositions as inputs and constrains the disentangled primitive features to be general across these compositions. More specifically, we leverage a compositional graph to define the overall primitive-sharing relationships between compositions, and build a task-specific architecture upon the recently successful large pre-trained vision-language model (VLM) CLIP, with dual cross-composition disentangling adapters (called L-Adapter and V-Adapter) inserted into CLIP's frozen text and image encoders, respectively. Evaluation on three popular CZSL benchmarks shows that our proposed solution significantly improves the performance of CZSL, and its components have been verified by solid ablation studies. Our code and data are available at:https://github.com/zhurunkai/DCDA.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes