ACTIS: improving data efficiency by leveraging semi-supervised augmentation consistency training for instance segmentation

Item Type:	Conference or Workshop Item
Title:	ACTIS: improving data efficiency by leveraging semi-supervised augmentation consistency training for instance segmentation
Creators Name:	Rumberger, J.L. and Franzen, J. and Hirsch, P. and Albrecht, J.P. and Kainmueller, D.
Abstract:	Segmenting objects like cells or nuclei in biomedical microscopy data is a standard task required for many downstream analyses. However, existing pre-trained models are continuously challenged by ever-evolving experimental setups and imaging platforms. On the other hand, training new models still requires a considerable number of annotated samples, rendering it infeasible for small to midsized experiments. To address this challenge, we propose a semi-supervised learning approach for instance segmentation that leverages a small number of annotated samples together with a larger number of unannotated samples. Our pipeline, Augmentation Consistency Training for Instance Segmentation (ACTIS), incorporates methods from consistency regularization and entropy minimization. In addition, we introduce a robust confidence-based loss masking scheme which we find empirically to work well on highly imbalanced class frequencies. We show that our model can surpass the performance of supervised models trained on more than twice as much annotated data. It achieves state-of-the-art results on three benchmark datasets in the biomedical domain, demonstrating its effectiveness for semi-supervised instance segmentation.
Keywords:	Instance Segmentation, Semi-supervised Training, Improve Data Efficiency, Benchmark Datasets, Semi-supervised Learning, Minimum Entropy, Biomedical Domain, Consistency Regularization, Semi-supervised Learning Approach, Denoising, Prediction Error, Validation Dataset, Data Augmentation, Teacher Model, Intersection over Union, Confidence Score, Semantic Segmentation, Additive Gaussian Noise, Matrix M, Affine Transformation, Semi-supervised Model, Pseudo Labels, Student Model, Self-supervised Learning, View Data, Auxiliary Task, Watershed Segmentation, Image X, Mouse Datasets, Considerable Improvement
Source:	2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)
ISSN:	2473-9944
ISBN:	979-8-3503-0744-3
Publisher:	IEEE
Page Range:	3792-3801
Date:	25 December 2023
Official Publication:	https://doi.org/10.1109/iccvw60793.2023.00410

Repository Staff Only: item control page