Sparse Object-level Supervision for Instance Segmentation with Pixel Embeddings

03/26/2021
by   Adrian Wolny, et al.
11

Most state-of-the-art instance segmentation methods have to be trained on densely annotated images. While difficult in general, this requirement is especially daunting for biomedical images, where domain expertise is often required for annotation. We propose to address the dense annotation bottleneck by introducing a proposal-free segmentation approach based on non-spatial embeddings, which exploits the structure of the learned embedding space to extract individual instances in a differentiable way. The segmentation loss can then be applied directly on the instances and the overall method can be trained on ground truth images where only a few objects are annotated, from scratch or in a semi-supervised transfer learning setting. In addition to the segmentation loss, our setup allows to apply self-supervised consistency losses on the unlabeled parts of the training data. We evaluate the proposed method on challenging 2D and 3D segmentation problems in different microscopy modalities as well as on the popular CVPPP instance segmentation benchmark where we achieve state-of-the-art results. The code is available at: https://github.com/kreshuklab/spoco

READ FULL TEXT

page 3

page 6

page 8

page 12

research
04/06/2022

Instance Segmentation of Unlabeled Modalities via Cyclic Segmentation GAN

Instance segmentation for unlabeled imaging modalities is a challenging ...
research
07/13/2023

AnyStar: Domain randomized universal star-convex 3D instance segmentation

Star-convex shapes arise across bio-microscopy and radiology in the form...
research
04/10/2019

Instance Segmentation of Biological Images Using Harmonic Embeddings

We present a new instance segmentation approach tailored to biological i...
research
07/06/2021

Stateless actor-critic for instance segmentation with high-level priors

Instance segmentation is an important computer vision problem which rema...
research
02/01/2018

Annotation-Free and One-Shot Learning for Instance Segmentation of Homogeneous Object Clusters

We propose a novel approach for instance segmen- tation given an image o...
research
07/12/2016

Gland Instance Segmentation by Deep Multichannel Side Supervision

In this paper, we propose a new image instance segmentation method that ...
research
04/21/2020

Instance Segmentation of Biomedical Images with an Object-aware Embedding Learned with Local Constraints

Automatic instance segmentation is a problem that occurs in many biomedi...

Please sign up or login with your details

Forgot password? Click here to reset