A Performance Increment Strategy for Semantic Segmentation of Low-Resolution Images from Damaged Roads

Rafael S. Toledo, Cristiano S. Oliveira, Vitor H. T. Oliveira, Eric A. Antonelo, Aldo von Wangenheim
11/25/2024
cs.CV

No official code found. Best implementation: tldrafael/pisss (2 stars, MIT, last commit 2024-11-27).

Abstract

Autonomous driving needs good roads, but 85% of Brazilian roads have damages that deep learning models may not regard as most semantic segmentation datasets for autonomous driving are high-resolution images of well-maintained urban roads. A representative dataset for emerging countries consists of low-resolution images of poorly maintained roads and includes labels of damage classes; in this scenario, three challenges arise: objects with few pixels, objects with undefined shapes, and highly underrepresented classes. To tackle these challenges, this work proposes the Performance Increment Strategy for Semantic Segmentation (PISSS) as a methodology of 14 training experiments to boost performance. With PISSS, we reached state-of-the-art results of 79.8 and 68.8 mIoU on the Road Traversing Knowledge (RTK) and Technik Autonomer Systeme 500 (TAS500) test sets, respectively. Furthermore, we also offer an analysis of DeepLabV3+ pitfalls for small object segmentation.

View on arXivView PDF

Code Implementations(1)

tldrafael/pisssCommunity100%
MIT

Performance Increment Strategy for Semantic Segmentation

20Python, MakefileSep 13, 20221 years agoMIT

Similar Papers

Solving Scene Understanding for Autonomous Navigation in Unstructured Environments

87% match2025

UdeerLID+: Integrating LiDAR, Image, and Relative Depth with Semi-Supervised

86% match2024

Applying Unsupervised Semantic Segmentation to High-Resolution UAV Imagery for Enhanced Road Scene Parsing

85% match1 repo62024

Progressive Query Refinement Framework for Bird's-Eye-View Semantic Segmentation from Surrounding Images

84% match1 repo22024

ANNA: A Deep Learning Based Dataset in Heterogeneous Traffic for Autonomous Vehicles

84% match2024

AD-SAM: Fine-Tuning the Segment Anything Vision Foundation Model for Autonomous Driving Perception

84% match2025

Cite this paper

@article{toledo2024performance,
  title  = {A Performance Increment Strategy for Semantic Segmentation of Low-Resolution Images from Damaged Roads},
  author = {Rafael S. Toledo and Cristiano S. Oliveira and Vitor H. T. Oliveira and Eric A. Antonelo and Aldo von Wangenheim},
  year   = {2024},
  eprint = {2411.16295},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2411.16295v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2411.16295.svg)](https://codeofpaper.com/paper/2411.16295)

Discussion