# Soft labelling for semantic segmentation: Bringing coherence to label down-sampling

Source: [Soft labelling for semantic segmentation: Bringing coherence to label down-sampling](<https://www.sota2.com/research/paper/68fa72371e209131cea55036>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T03:50:18.590Z

Authors: Roberto Alcover-Couso, Marcos Escudero-Vinolo, Juan C. SanMiguel, Jose M. Martinez

Published: 2023-02-27T17:02:30.000Z

## Abstract

In semantic segmentation, training data down-sampling is commonly performed due to limited resources, the need to adapt image size to the model input, or improve data augmentation. This down-sampling typically employs different strategies for the image data and the annotated labels. Such discrepancy leads to mismatches between the down-sampled color and label images. Hence, the training performance significantly decreases as the down-sampling factor increases. In this paper, we bring together the down-sampling strategies for the image data and the training labels. To that aim, we propose a novel framework for label down-sampling via soft-labeling that better conserves label information after down-sampling. Therefore, fully aligning soft-labels with image data to keep the distribution of the sampled pixels. This proposal also produces reliable annotations for under-represented semantic classes. Altogether, it allows training competitive models at lower resolutions. Experiments show that the proposal outperforms other down-sampling strategies. Moreover, state-of-the-art performance is achieved for reference benchmarks, but employing significantly less computational resources than foremost approaches. This proposal enables competitive research for semantic segmentation under resource constraints.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Semantic Segmentation on Mapillary (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-mapillary-val.md>): Semantic segmentation; Mapillary (val); mIoU: 54.4.

- [Semantic Segmentation on ADE20K (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-val.md>): Semantic segmentation; ADE20K (val); mIoU: 48.3.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2302.13961>)

[PDF](<https://arxiv.org/pdf/2302.13961.pdf>)