CobNet: cross attention on object and background for few-shot segmentation

TY - CHAP

T1 - CobNet: cross attention on object and background for few-shot segmentation

AU - Guan, Haoyan

AU - Spratling, Michael

N1 - Publisher Copyright: © 2022 IEEE.

PY - 2022/11/29

Y1 - 2022/11/29

N2 - Few-shot segmentation aims to segment images containing objects from previously unseen classes using only a few annotated samples. Most current methods focus on using object information extracted, with the aid of human annotations, from support images to identify the same objects in new query images. However, background information can also be useful to distinguish objects from their surroundings. Hence, some previous methods also extract background information from the support images. In this paper, we argue that such information is of limited utility, as the background in different images can vary widely. To overcome this issue, we propose CobNet which utilises information about the background that is extracted from the query images without annotations of those images. Experiments show that our method achieves a mean Intersection-over-Union score of 61.4% and 37.8% for 1-shot segmentation on PASCAL-5i and COCO-20i respectively, outperforming previous methods. It is also shown to produce state-of-the-art performances of 53.7% for weakly-supervised few-shot segmentation, where no annotations are provided for the support images.

AB - Few-shot segmentation aims to segment images containing objects from previously unseen classes using only a few annotated samples. Most current methods focus on using object information extracted, with the aid of human annotations, from support images to identify the same objects in new query images. However, background information can also be useful to distinguish objects from their surroundings. Hence, some previous methods also extract background information from the support images. In this paper, we argue that such information is of limited utility, as the background in different images can vary widely. To overcome this issue, we propose CobNet which utilises information about the background that is extracted from the query images without annotations of those images. Experiments show that our method achieves a mean Intersection-over-Union score of 61.4% and 37.8% for 1-shot segmentation on PASCAL-5i and COCO-20i respectively, outperforming previous methods. It is also shown to produce state-of-the-art performances of 53.7% for weakly-supervised few-shot segmentation, where no annotations are provided for the support images.

UR - http://www.scopus.com/inward/record.url?scp=85143608920&partnerID=8YFLogxK

U2 - 10.1109/ICPR56361.2022.9956070

DO - 10.1109/ICPR56361.2022.9956070

M3 - Conference paper

T3 - Proceedings - International Conference on Pattern Recognition

SP - 39

EP - 45

BT - 2022 26th International Conference on Pattern Recognition, ICPR 2022

T2 - International Conference on Pattern Recognition

Y2 - 22 August 2022 through 25 August 2022

ER -