Depth Edge Alignment Loss: DEALing with Depth in Weakly Supervised Semantic Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schmidt, Patrick, Belagiannis, Vasileios, Nalpantidis, Lazaros
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908766181523456
author Schmidt, Patrick
Belagiannis, Vasileios
Nalpantidis, Lazaros
author_facet Schmidt, Patrick
Belagiannis, Vasileios
Nalpantidis, Lazaros
contents Autonomous robotic systems applied to new domains require an abundance of expensive, pixel-level dense labels to train robust semantic segmentation models under full supervision. This study proposes a model-agnostic Depth Edge Alignment Loss to improve Weakly Supervised Semantic Segmentation models across different datasets. The methodology generates pixel-level semantic labels from image-level supervision, avoiding expensive annotation processes. While weak supervision is widely explored in traditional computer vision, our approach adds supervision with pixel-level depth information, a modality commonly available in robotic systems. We demonstrate how our approach improves segmentation performance across datasets and models, but can also be combined with other losses for even better performance, with improvements up to +5.439, +1.274 and +16.416 points in mean Intersection over Union on the PASCAL VOC / MS COCO validation, and the HOPE static onboarding split, respectively. Our code is made publicly available at https://github.com/DTU-PAS/DEAL.
format Preprint
id arxiv_https___arxiv_org_abs_2509_17702
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Depth Edge Alignment Loss: DEALing with Depth in Weakly Supervised Semantic Segmentation
Schmidt, Patrick
Belagiannis, Vasileios
Nalpantidis, Lazaros
Computer Vision and Pattern Recognition
Autonomous robotic systems applied to new domains require an abundance of expensive, pixel-level dense labels to train robust semantic segmentation models under full supervision. This study proposes a model-agnostic Depth Edge Alignment Loss to improve Weakly Supervised Semantic Segmentation models across different datasets. The methodology generates pixel-level semantic labels from image-level supervision, avoiding expensive annotation processes. While weak supervision is widely explored in traditional computer vision, our approach adds supervision with pixel-level depth information, a modality commonly available in robotic systems. We demonstrate how our approach improves segmentation performance across datasets and models, but can also be combined with other losses for even better performance, with improvements up to +5.439, +1.274 and +16.416 points in mean Intersection over Union on the PASCAL VOC / MS COCO validation, and the HOPE static onboarding split, respectively. Our code is made publicly available at https://github.com/DTU-PAS/DEAL.
title Depth Edge Alignment Loss: DEALing with Depth in Weakly Supervised Semantic Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.17702