Few-shot Semantic Segmentation with Self-supervision from Pseudo-classes

2021-10-22 · Yiwen Li, Gratianus Wesley Putra Data, Yunguan Fu, Yipeng Hu, Victor Adrian Prisacariu

Despite the success of deep learning methods for semantic segmentation, few-shot semantic segmentation remains a challenging task due to the limited training data and the generalisation requirement for unseen classes. While recent progress has been particularly encouraging, we discover that existing methods tend to have poor performance in terms of meanIoU when query images contain other semantic classes besides the target class. To address this issue, we propose a novel self-supervised task that generates random pseudo-classes in the background of the query images, providing extra training data that would otherwise be unavailable when predicting individual target classes. To that end, we adopted superpixel segmentation for generating the pseudo-classes. With this extra supervision, we improved the meanIoU performance of the state-of-the-art method by 2.5% and 5.1% on the one-shot tasks, as well as 6.7% and 4.4% on the five-shot tasks, on the PASCAL-5i and COCO benchmarks, respectively.

📄 PDF Abstract BibTeX arXiv:2110.11742

Code (1)

kate-sann5100/SS_few_shot 공식 구현 pytorch

Tasks

Few-Shot Semantic Segmentation Segmentation Semantic Segmentation

Self-Supervision with Superpixels: Training Few-shot Medical Image Segmentation without Annotation

2020-07-20 · ECCV 2020 8 · Cheng Ouyang, Carlo Biffi, Chen Chen, Turkay Kart 외

Few-shot semantic segmentation (FSS) has great potential for medical imaging applications. Most of the existing FSS techniques require abundant annotated semantic classes for training. However, these methods may not be a…

Cardiac Segmentation Few-Shot Semantic Segmentation Image Segmentation Medical Image Segmentation+4

Distilling Self-Supervised Vision Transformers for Weakly-Supervised Few-Shot Classification & Segmentation

2023-07-07 · CVPR 2023 1 · Dahyun Kang, Piotr Koniusz, Minsu Cho, Naila Murray

We address the task of weakly-supervised few-shot image classification and segmentation, by leveraging a Vision Transformer (ViT) pretrained with self-supervision. Our proposed method takes token representations from the…

Few-Shot Image Classification image-classification Image Classification Pseudo Label+1

SIGN: Spatial-information Incorporated Generative Network for Generalized Zero-shot Semantic Segmentation

2021-08-27 · ICCV 2021 10 · Jiaxin Cheng, Soumyaroop Nandi, Prem Natarajan, Wael Abd-Almageed

Unlike conventional zero-shot classification, zero-shot semantic segmentation predicts a class label at the pixel level instead of the image level. When solving zero-shot semantic segmentation problems, the need for pixe…

Knowledge Distillation Segmentation Semantic Segmentation zero-shot-classification+2

Recursive Training for Zero-Shot Semantic Segmentation

2021-02-26 · Ce Wang, Moshiur Farazi, Nick Barnes

General purpose semantic segmentation relies on a backbone CNN network to extract discriminative features that help classify each image pixel into a 'seen' object class (ie., the object classes available during training)…

Segmentation Semantic Segmentation Zero-Shot Semantic Segmentation

Semi-supervision semantic segmentation with uncertainty-guided self cross supervision

2022-03-10 · Yunyang Zhang, Zhiqiang Gong, Xiaohu Zheng, Xiaoyu Zhao 외

As a powerful way of realizing semi-supervised segmentation, the cross supervision method learns cross consistency based on independent ensemble models using abundant unlabeled images. However, the wrong pseudo labeling …

Segmentation Semantic Segmentation