Adaptive Early-Learning Correction for Segmentation from Noisy Annotations

by   Sheng Liu, et al.

Deep learning in the presence of noisy annotations has been studied extensively in classification, but much less in segmentation tasks. In this work, we study the learning dynamics of deep segmentation networks trained on inaccurately-annotated data. We discover a phenomenon that has been previously reported in the context of classification: the networks tend to first fit the clean pixel-level labels during an "early-learning" phase, before eventually memorizing the false annotations. However, in contrast to classification, memorization in segmentation does not arise simultaneously for all semantic categories. Inspired by these findings, we propose a new method for segmentation from noisy annotations with two key elements. First, we detect the beginning of the memorization phase separately for each category during training. This allows us to adaptively correct the noisy annotations in order to exploit early learning. Second, we incorporate a regularization term that enforces consistency across scales to boost robustness against annotation noise. Our method outperforms standard approaches on a medical-imaging segmentation task where noises are synthesized to mimic human annotation errors. It also provides robustness to realistic noisy annotations present in weakly-supervised semantic segmentation, achieving state-of-the-art results on PASCAL VOC 2012.


page 4

page 9

page 19

page 20

page 21

page 22

page 23


Early-Learning Regularization Prevents Memorization of Noisy Labels

We propose a novel framework to perform classification via deep learning...

AutoCorrect: Deep Inductive Alignment of Noisy Geometric Annotations

We propose AutoCorrect, a method to automatically learn object-annotatio...

Clearing noisy annotations for computed tomography imaging

One of the problems on the way to successful implementation of neural ne...

Learning to Segment Skin Lesions from Noisy Annotations

Deep convolutional neural networks have driven substantial advancements ...

Learning Pixel-wise Labeling from the Internet without Human Interaction

Deep learning stands at the forefront in many computer vision tasks. How...

Weakly Supervised Segmentation by A Deep Geodesic Prior

The performance of the state-of-the-art image segmentation methods heavi...

Pseudo-Label Ensemble-based Semi-supervised Learning for Handling Noisy Soiling Segmentation Annotations

Manual annotation of soiling on surround view cameras is a very challeng...

Please sign up or login with your details

Forgot password? Click here to reset