TY - GEN
T1 - A Strong Baseline for Generalized Few-Shot Semantic Segmentation
AU - Hajimiri, Sina
AU - Boudiaf, Malik
AU - Ayed, Ismail Ben
AU - Dolz, Jose
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - This paper introduces a generalized few-shot segmentation framework with a straightforward training process and an easy-to-optimize inference phase. In particular, we propose a simple yet effective model based on the well-known InfoMax principle, where the Mutual Information (MI) between the learned feature representations and their corresponding predictions is maximized. In addition, the terms derived from our MI-based formulation are coupled with a knowledge distillation term to retain the knowledge on base classes. With a simple training process, our inference model can be applied on top of any segmentation network trained on base classes. The proposed inference yields substantial improvements on the popular few-shot segmentation benchmarks, PASCAL-5i and COCO-20i. Particularly, for novel classes, the improvement gains range from 7% to 26% (PASCAL-5i) and from 3% to 12% (COCO-20i) in the 1-shot and 5-shot scenarios, respectively. Furthermore, we propose a more challenging setting, where performance gaps are further exacerbated. Our code is publicly available at https://github.com/sinahmr/DIaM.
AB - This paper introduces a generalized few-shot segmentation framework with a straightforward training process and an easy-to-optimize inference phase. In particular, we propose a simple yet effective model based on the well-known InfoMax principle, where the Mutual Information (MI) between the learned feature representations and their corresponding predictions is maximized. In addition, the terms derived from our MI-based formulation are coupled with a knowledge distillation term to retain the knowledge on base classes. With a simple training process, our inference model can be applied on top of any segmentation network trained on base classes. The proposed inference yields substantial improvements on the popular few-shot segmentation benchmarks, PASCAL-5i and COCO-20i. Particularly, for novel classes, the improvement gains range from 7% to 26% (PASCAL-5i) and from 3% to 12% (COCO-20i) in the 1-shot and 5-shot scenarios, respectively. Furthermore, we propose a more challenging setting, where performance gaps are further exacerbated. Our code is publicly available at https://github.com/sinahmr/DIaM.
KW - Transfer
KW - continual
KW - low-shot
KW - meta
KW - or long-tail learning
UR - https://www.scopus.com/pages/publications/85165673782
U2 - 10.1109/CVPR52729.2023.01084
DO - 10.1109/CVPR52729.2023.01084
M3 - Contribution to conference proceedings
AN - SCOPUS:85165673782
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 11269
EP - 11278
BT - Proceedings - 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023
PB - IEEE Computer Society
T2 - 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023
Y2 - 17 June 2023 through 24 June 2023
ER -