# Meta-Learning Initializations for Image Segmentation

arXiv: https://arxiv.org/abs/1912.06290 · HTML: https://mlsanity.com/paper/1912.06290/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Dec 2019.

## Built on by (most cited first, 12 of 12)

- [Few-shot Segmentation with Optimal Transport Matching and Message Flow](https://mlsanity.com/paper/2108.08518/) (2021)
- [Interclass Prototype Relation for Few-Shot Segmentation](https://mlsanity.com/paper/2211.08681/) (2022)
- [CRCNet: Few-shot Segmentation with Cross-Reference and Region-Global Conditional Networks](https://mlsanity.com/paper/2208.10761/) (2022)
- [PatchRefineNet: Improving Binary Segmentation by Incorporating Signals from Optimal Patch-wise Binarization](https://mlsanity.com/paper/2211.06560/) (2022)
- [Slide-SAM: Medical SAM Meets Sliding Window](https://mlsanity.com/paper/2311.10121/) (2023)
- [A Meta-Learning Approach for Medical Image Registration](https://mlsanity.com/paper/2104.10447/) (2021)
- [Few Shot Semantic Segmentation: a review of methodologies, benchmarks, and open challenges](https://mlsanity.com/paper/2304.05832/) (2023)
- [Meta-Learners for Few-Shot Weakly-Supervised Medical Image Segmentation](https://mlsanity.com/paper/2305.06912/) (2023)
- [Unsupervised augmentation optimization for few-shot medical image segmentation](https://mlsanity.com/paper/2306.05107/) (2023)
- [Meta-Decomposition: Dynamic Segmentation Approach Selection in IoT-based Activity Recognition](https://mlsanity.com/paper/2404.11742/) (2024)
- [Perturbing the Gradient for Alleviating Meta Overfitting](https://mlsanity.com/paper/2405.12299/) (2024)
- [Meta-Learning for Classifier Selection in Image Datasets: A Feature-Driven Framework for Accuracy Prediction](https://mlsanity.com/paper/2609.11041/) (2026)

## Abstract (arXiv)

We extend first-order model agnostic meta-learning algorithms (including FOMAML and Reptile) to image segmentation, present a novel neural network architecture built for fast learning which we call EfficientLab, and leverage a formal definition of the test error of meta-learning algorithms to decrease error on out of distribution tasks. We show state of the art results on the FSS-1000 dataset by meta-training EfficientLab with FOMAML and using Bayesian optimization to infer the optimal test-time adaptation routine hyperparameters. We also construct a small benchmark dataset, FP-k, for the empirical study of how meta-learning systems perform in both few- and many-shot settings. On the FP-k dataset, we show that meta-learned initializations provide value for canonical few-shot image segmentation but their performance is quickly matched by conventional transfer learning with performance being equal beyond 10 labeled examples. Our code, meta-learned model, and the FP-k dataset are available at https://github.com/ml4ai/mliis .