Single-Stage Semantic Segmentation From Image Labels

By Nikita Araslanov, Stefan Roth. 2020

335 citations recorded by the source corpus.

Recent years have seen a rapid growth in new approaches improving the accuracy of semantic segmentation in a weakly supervised setting, i.e. with only image-level labels available for training. However, this has come at the cost of increased model complexity and sophisticated multi-stage training procedures. This is in contrast to earlier work that used only a single stage -- training one segmentation network on image labels -- which was abandoned due to inferior segmentation accuracy. In this work, we first define three desirable properties of a weakly supervised method: local consistency, semantic fidelity, and completeness. Using these properties as guidelines, we then develop a segmentation-based network model and a self-supervised training scheme to train for semantic masks from image-level annotations in a single stage. We show that despite its simplicity, our method achieves results that are competitive with significantly more complex pipelines, substantially outperforming earlier single-stage methods.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Incremental Learning in Semantic Segmentation from Image Labels (2021, methodology): “Directly applying existing weakly-supervised methods to incremental segmentation would require to (i) extract pixel-wise pseudo-supervision offline using a weakly supervised approach [3, 5, 36, 59, 63] and (ii) update the segmentation network resorting to an incremental learning technique [8,21,43].”
Railroad is not a Train: Saliency as Pseudo-pixel Supervision for Weakly Supervised Semantic Segmentation (2021, methodology): “Recently, [3] uses a single segmentation-based model with a self-supervised training scheme.”
LayerCAM: Exploring Hierarchical Class Activation Maps for Localization (2021, background): “Some works [71], [72], [73] utilize image labels to train segmentation models directly.”