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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Geng, Yuxia, Zhu, Runkai, Chen, Jiaoyan, Chen, Jintai, Chen, Xiang, Chen, Zhuo, Qiao, Shuofei, Wang, Yuxiang, Xu, Xiaoliang, Huang, Sheng-Jun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908384520830976
author Geng, Yuxia
Zhu, Runkai
Chen, Jiaoyan
Chen, Jintai
Chen, Xiang
Chen, Zhuo
Qiao, Shuofei
Wang, Yuxiang
Xu, Xiaoliang
Huang, Sheng-Jun
author_facet Geng, Yuxia
Zhu, Runkai
Chen, Jiaoyan
Chen, Jintai
Chen, Xiang
Chen, Zhuo
Qiao, Shuofei
Wang, Yuxiang
Xu, Xiaoliang
Huang, Sheng-Jun
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2408_09786
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Graph-guided Cross-composition Feature Disentanglement for Compositional Zero-shot Learning
Geng, Yuxia
Zhu, Runkai
Chen, Jiaoyan
Chen, Jintai
Chen, Xiang
Chen, Zhuo
Qiao, Shuofei
Wang, Yuxiang
Xu, Xiaoliang
Huang, Sheng-Jun
Computer Vision and Pattern Recognition
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.
title Graph-guided Cross-composition Feature Disentanglement for Compositional Zero-shot Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.09786