Dense Gaussian Processes for Few-Shot Segmentation

10/07/2021
by   Joakim Johnander, et al.
0

Few-shot segmentation is a challenging dense prediction task, which entails segmenting a novel query image given only a small annotated support set. The key problem is thus to design a method that aggregates detailed information from the support set, while being robust to large variations in appearance and context. To this end, we propose a few-shot segmentation method based on dense Gaussian process (GP) regression. Given the support set, our dense GP learns the mapping from local deep image features to mask values, capable of capturing complex appearance distributions. Furthermore, it provides a principled means of capturing uncertainty, which serves as another powerful cue for the final segmentation, obtained by a CNN decoder. Instead of a one-dimensional mask output, we further exploit the end-to-end learning capabilities of our approach to learn a high-dimensional output space for the GP. Our approach sets a new state-of-the-art for both 1-shot and 5-shot FSS on the PASCAL-5^i and COCO-20^i benchmarks, achieving an absolute gain of +14.9 mIoU in the COCO-20^i 5-shot setting. Furthermore, the segmentation quality of our approach scales gracefully when increasing the support set size, while achieving robust cross-dataset transfer.

READ FULL TEXT

page 3

page 9

page 19

page 20

page 21

research
03/30/2021

Deep Gaussian Processes for Few-Shot Segmentation

Few-shot segmentation is a challenging task, requiring the extraction of...
research
08/11/2021

Few-Shot Segmentation with Global and Local Contrastive Learning

In this work, we address the challenging task of few-shot segmentation. ...
research
08/23/2022

CRCNet: Few-shot Segmentation with Cross-Reference and Region-Global Conditional Networks

Few-shot segmentation aims to learn a segmentation model that can be gen...
research
12/09/2022

MSI: Maximize Support-Set Information for Few-Shot Segmentation

FSS(Few-shot segmentation) aims to segment a target class with a small n...
research
03/15/2020

DeepEMD: Few-Shot Image Classification with Differentiable Earth Mover's Distance and Structured Classifiers

In this paper, we address the few-shot classification task from a new pe...
research
08/19/2021

Few-shot Segmentation with Optimal Transport Matching and Message Flow

We address the challenging task of few-shot segmentation in this work. I...
research
03/25/2020

Learning What to Learn for Video Object Segmentation

Video object segmentation (VOS) is a highly challenging problem, since t...

Please sign up or login with your details

Forgot password? Click here to reset