Fine-grained Few-shot Recognition by Deep Object Parsing
Ruizhao Zhu (Boston University),* Pengkai Zhu (Amazon Web Services), Samarth Mishra (Boston University), Venkatesh Saligrama (Boston University)
The 34th British Machine Vision Conference
Abstract
We presented DOP, a deep object-parsing method for fine-grained few-shot recognition. Our fundamental concept is that, while different object classes exhibit novel visual appearance, at a sufficiently small scale, visual patterns are duplicated. Hence, by leveraging training data to learn a dictionary of templates distributed across different relative locations, an object can be recognized simply by identifying which of the templates in the dictionary are expressed, and how these patterns are geometrically distributed. We build a statistical model for parsing that takes the output of a convolutional backbone as input to produce a parsed output. We then post-hoc learn to re-weight query and support instances to identify the best matching class, and as such this procedure allows for mitigating visual distortions. Our proposed method is an end-to-end deep neural network training method, and we show that our performance is not only competitive but also the outputs generated are interpretable.Video
Citation
@inproceedings{Zhu_2023_BMVC,
author = {Ruizhao Zhu and Pengkai Zhu and Samarth Mishra and Venkatesh Saligrama},
title = {Fine-grained Few-shot Recognition by Deep Object Parsing},
booktitle = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen, UK, November 20-24, 2023},
publisher = {BMVA},
year = {2023},
url = {https://papers.bmvc2023.org/0330.pdf}
}
Copyright © 2023
The British Machine Vision Conference is organised by