Saved in:
Bibliographic Details
Main Authors: Ke, Shuyan, Mei, Yifan, Wu, Changli, Zheng, Yonghan, Ji, Jiayi, Cao, Liujuan, Ji, Rongrong
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.15670
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914482303795200
author Ke, Shuyan
Mei, Yifan
Wu, Changli
Zheng, Yonghan
Ji, Jiayi
Cao, Liujuan
Ji, Rongrong
author_facet Ke, Shuyan
Mei, Yifan
Wu, Changli
Zheng, Yonghan
Ji, Jiayi
Cao, Liujuan
Ji, Rongrong
contents Reasoning segmentation has recently expanded from ground-level scenes to remote-sensing imagery, yet UAV data poses distinct challenges, including oblique viewpoints, ultra-high resolutions, and extreme scale variations. To address these issues, we formally define the UAV Reasoning Segmentation task and organize its semantic requirements into three dimensions: Spatial, Attribute, and Scene-level reasoning. Based on this formulation, we construct DRSeg, a large-scale benchmark for UAV reasoning segmentation, containing 10k high-resolution aerial images paired with Chain-of-Thought QA supervision across all three reasoning types. As a benchmark companion, we introduce PixDLM, a simple yet effective pixel-level multimodal language model that serves as a unified baseline for this task. Experiments on DRSeg establish strong baseline results and highlight the unique challenges of UAV reasoning segmentation, providing a solid foundation for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2604_15670
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PixDLM: A Dual-Path Multimodal Language Model for UAV Reasoning Segmentation
Ke, Shuyan
Mei, Yifan
Wu, Changli
Zheng, Yonghan
Ji, Jiayi
Cao, Liujuan
Ji, Rongrong
Computer Vision and Pattern Recognition
Reasoning segmentation has recently expanded from ground-level scenes to remote-sensing imagery, yet UAV data poses distinct challenges, including oblique viewpoints, ultra-high resolutions, and extreme scale variations. To address these issues, we formally define the UAV Reasoning Segmentation task and organize its semantic requirements into three dimensions: Spatial, Attribute, and Scene-level reasoning. Based on this formulation, we construct DRSeg, a large-scale benchmark for UAV reasoning segmentation, containing 10k high-resolution aerial images paired with Chain-of-Thought QA supervision across all three reasoning types. As a benchmark companion, we introduce PixDLM, a simple yet effective pixel-level multimodal language model that serves as a unified baseline for this task. Experiments on DRSeg establish strong baseline results and highlight the unique challenges of UAV reasoning segmentation, providing a solid foundation for future research.
title PixDLM: A Dual-Path Multimodal Language Model for UAV Reasoning Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.15670