Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning
Abstract
Learning from seen attribute-object pairs to generalize to unseen compositions has been studied extensively in Compositional Zero-Shot Learning (CZSL). However CZSL setup is still limited to seen attributes and objects and cannot generalize to unseen concepts and their compositions. To overcome this limitation we propose a new task Open Vocabulary-Compositional Zero-shot Learning (OV-CZSL) where unseen attributes objects and unseen compositions are evaluated. To show that OV-CZSL is a challenging yet solvable problem we propose three new benchmarks based on existing datasets MIT-States C-GQA and VAW-CZSL along with new baselines and evaluation setup. We use language embeddings and external vocabulary with our novel neighborhood expansion loss to allow any method to learn semantic correlations between seen and unseen primitives.
Cite
Text
Saini et al. "Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning." Conference on Computer Vision and Pattern Recognition, 2024. doi:10.1109/CVPR52733.2024.01371Markdown
[Saini et al. "Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning." Conference on Computer Vision and Pattern Recognition, 2024.](https://mlanthology.org/cvpr/2024/saini2024cvpr-beyond/) doi:10.1109/CVPR52733.2024.01371BibTeX
@inproceedings{saini2024cvpr-beyond,
title = {{Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning}},
author = {Saini, Nirat and Pham, Khoi and Shrivastava, Abhinav},
booktitle = {Conference on Computer Vision and Pattern Recognition},
year = {2024},
pages = {14466-14476},
doi = {10.1109/CVPR52733.2024.01371},
url = {https://mlanthology.org/cvpr/2024/saini2024cvpr-beyond/}
}