Efficient Pothole Segmentation Using Quantization and Pruning for ADAS
Revista de Informática Teórica e Aplicada · 10.03.2026 · 33 (2) · 302-309
DOI: 10.22456/2175-2745.150751 ↗Kurz gefasst
Fachartikel in der Zeitschrift Revista de Informática Teórica e Aplicada, online erschienen am 10.03.2026. Verfasst von William D'Abruzzo Martins und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Real-time semantic segmentation is critical for Advanced Driver-Assistance Systems (ADAS), but deploying deep learning models on resource-constrained hardware remains a challenge. This work presents an efficient pipeline for pothole segmentation using the lightweight PP-LiteSeg model, evaluating and comparing two key optimization techniques: post-training quantization and network pruning. Results demonstrate that 8-bit integer (INT8) quantization is highly effective, achieving a 3.54x speedup in GPU compute time with only a minimal, well-controlled loss in segmentation accuracy. While network pruning also reduces model size and latency, it comes at a more pronounced cost to performance. In conclusion, while quantization offers a superior trade-off between speed and accuracy, network pruning remains a valuable strategy for applications where memory compression is the primary objective.
Zitieren
APA
Martins, W. D., Osório, F. S., & Bruno, D. R. (2026). Efficient Pothole Segmentation Using Quantization and Pruning for ADAS. Revista de Informática Teórica e Aplicada, 33(2), 302-309. https://doi.org/10.22456/2175-2745.150751
Vancouver
Martins WD, Osório FS, Bruno DR. Efficient Pothole Segmentation Using Quantization and Pruning for ADAS. Revista de Informática Teórica e Aplicada. 2026;33(2):302-309. doi:10.22456/2175-2745.150751
BibTeX
@article{Martins2026,
author = {Martins, William D'Abruzzo and Osório, Fernando Santos and Bruno, Diego Renan},
title = {Efficient Pothole Segmentation Using Quantization and Pruning for ADAS},
journal = {Revista de Informática Teórica e Aplicada},
year = {2026},
volume = {33},
number = {2},
pages = {302--309},
doi = {10.22456/2175-2745.150751}
}
EUVIMED