Hyperbolic Active Learning for Semantic Segmentation under Domain Shift

by   Luca Franco, et al.

For the task of semantic segmentation (SS) under domain shift, active learning (AL) acquisition strategies based on image regions and pseudo labels are state-of-the-art (SoA). The presence of diverse pseudo-labels within a region identifies pixels between different classes, which is a labeling efficient active learning data acquisition strategy. However, by design, pseudo-label variations are limited to only select the contours of classes, limiting the final AL performance. We approach AL for SS in the Poincaré hyperbolic ball model for the first time and leverage the variations of the radii of pixel embeddings within regions as a novel data acquisition strategy. This stems from a novel geometric property of a hyperbolic space trained without enforced hierarchies, which we experimentally prove. Namely, classes are mapped into compact hyperbolic areas with a comparable intra-class radii variance, as the model places classes of increasing explainable difficulty at denser hyperbolic areas, i.e. closer to the Poincaré ball edge. The variation of pixel embedding radii identifies well the class contours, but they also select a few intra-class peculiar details, which boosts the final performance. Our proposed HALO (Hyperbolic Active Learning Optimization) surpasses the supervised learning performance for the first time in AL for SS under domain shift, by only using a small portion of labels (i.e., 1 experimental analysis is based on two established benchmarks, i.e. GTAV → Cityscapes and SYNTHIA → Cityscapes, where we set a new SoA. The code will be released.


page 2

page 6

page 16

page 17

page 18

page 19


Towards Fewer Annotations: Active Learning via Region Impurity and Prediction Uncertainty for Domain Adaptive Semantic Segmentation

Self-training has greatly facilitated domain adaptive semantic segmentat...

Active Learning for Semantic Segmentation with Multi-class Label Query

This paper proposes a new active learning method for semantic segmentati...

DEAL: Difficulty-aware Active Learning for Semantic Segmentation

Active learning aims to address the paucity of labeled data by finding t...

Mining of Single-Class by Active Learning for Semantic Segmentation

Several Active Learning (AL) policies require retraining a target model ...

Reducing Annotation Effort by Identifying and Labeling Contextually Diverse Classes for Semantic Segmentation Under Domain Shift

In Active Domain Adaptation (ADA), one uses Active Learning (AL) to sele...

Hyperbolic Uncertainty Aware Semantic Segmentation

Semantic segmentation (SS) aims to classify each pixel into one of the p...

Learning to Detect Semantic Boundaries with Image-level Class Labels

This paper presents the first attempt to learn semantic boundary detecti...

Please sign up or login with your details

Forgot password? Click here to reset