Saved in:
Bibliographic Details
Main Authors: Rajan, Shreshth, Liu, Raymond
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.18082
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915688079163392
author Rajan, Shreshth
Liu, Raymond
author_facet Rajan, Shreshth
Liu, Raymond
contents Semantic segmentation of outdoor street scenes plays a key role in applications such as autonomous driving, mobile robotics, and assistive technology for visually-impaired pedestrians. For these applications, accurately distinguishing between key surfaces and objects such as roads, sidewalks, vehicles, and pedestrians is essential for maintaining safety and minimizing risks. Semantic segmentation must be robust to different environments, lighting and weather conditions, and sensor noise, while being performed in real-time. We propose a region-level, uncertainty-gated retrieval mechanism that improves segmentation accuracy and calibration under domain shift. Our best method achieves an 11.3% increase in mean intersection-over-union while reducing retrieval cost by 87.5%, retrieving for only 12.5% of regions compared to 100% for always-on baseline.
format Preprint
id arxiv_https___arxiv_org_abs_2512_18082
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Uncertainty-Gated Region-Level Retrieval for Robust Semantic Segmentation
Rajan, Shreshth
Liu, Raymond
Computer Vision and Pattern Recognition
Artificial Intelligence
Semantic segmentation of outdoor street scenes plays a key role in applications such as autonomous driving, mobile robotics, and assistive technology for visually-impaired pedestrians. For these applications, accurately distinguishing between key surfaces and objects such as roads, sidewalks, vehicles, and pedestrians is essential for maintaining safety and minimizing risks. Semantic segmentation must be robust to different environments, lighting and weather conditions, and sensor noise, while being performed in real-time. We propose a region-level, uncertainty-gated retrieval mechanism that improves segmentation accuracy and calibration under domain shift. Our best method achieves an 11.3% increase in mean intersection-over-union while reducing retrieval cost by 87.5%, retrieving for only 12.5% of regions compared to 100% for always-on baseline.
title Uncertainty-Gated Region-Level Retrieval for Robust Semantic Segmentation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2512.18082