Skip to content

Hi-OPD: Hierarchy-Aware Open-Prompt Detection for Remote Sensing Images

Sep 2026 · arXiv.org · Vol abs/2609.25584 · 1 citation · 38 references
Computer Science

TL;DR

Joint AP and CAR show that explicit hierarchy training repairs this failure mode while retaining atomic detection and prompt transfer, and shows that explicit hierarchy training repairs this failure mode while retaining atomic detection and prompt transfer.

Abstract

Hi-OPD addresses a failure mode left uncontrolled by flat open-prompt training: descendant retrieval need not persist under ancestor queries when multi-source remote sensing annotations exhibit inconsistent granularity and missing labels. A detector may localize \textit{car} and \textit{van} under atomic prompts yet miss the same instances under \textit{vehicle}; flat AP does not expose this cross-level inconsistency. We propose Hi-OPD, a hierarchy-aware open-prompt detector, and construct RS153-HierOPD from 175,644 retained training image/tile records and 3.48M boxes mapped to 153 atomic categories with sparse hierarchy and alias relations. Hi-OPD learns ancestor retrieval through hierarchy-safe negative sampling, path multi-positive supervision, and one-way upward consistency, while per-source risk exclusion handles potentially missing labels. ConvVPE converts K-shot support boxes into text-compatible embeddings using detector-native features and the shared contrastive head. On Track A, Hi-OPD obtains 79.7/72.3 AP50 on DIOR/DOTA-v2.0, above the literature-reported OpenRSD results of 76.7/71.8. Under controlled training on the original converted annotations, the full hierarchy recipe raises DOTA-v2.0 parent AP50 from 7.2 to 71.5 and FAIR1M grandparent AP50 from 31.6 to 71.4, while DOTA-v2.0 atomic AP50 changes from 71.4 to 72.3. The text path reaches 99.7% CAR50 (0.3% violation) across the three common sources and 99.9%/0.1% on FAIR1M grandparent relations. On held-out VEDAI, text AP50 is 75.9, 6.2 points above OpenRSD. Joint AP and CAR show that explicit hierarchy training repairs this failure mode while retaining atomic detection and prompt transfer.

View source

Similar papers

Open access 2026

CRS-Det: Consistent Region Semantic Modeling for Remote Sensing Open-Vocabulary Object Detection

Remote sensing open-vocabulary object detection (RSOVD) aims to detect objects beyond the categories annotated during training. Although vision–language models (VLMs) have significantly advanced this task, existing approaches often suffer from region-level semantic instability, where region representations are highly s...

Ming Liu, Jing Li, Dong Ren et al. · 0 citations
Open access 2026

Dual-Level Prototype Alignment via Cross-Attention for Few-Shot Remote Sensing Semantic Segmentation

DLPANet is proposed, a novel dual-level prototype alignment network centered on Prototype-Guided Spatial Attention, enabling simultaneous modeling of scene context and fine-grained details and demonstrates that the decoupled dual cross-attention mechanism provides superior prototype-query alignment compared to prior gl...

Mustafa Alawadi, M. Fateh · 0 citations
Open access 2026

RS-ReXSeg: Rejection-Aware Explanatory Language-Guided Segmentation for Open-Query Remote Sensing Images

Language-guided remote sensing image segmentation enables pixel-level localization from natural-language queries, but most existing methods implicitly assume that the queried target is present. In open-query interpretation, absent targets and visually similar land-cover objects may cause models to generate incorrect ma...

Guang-Chang Li, Yu-Rong Qian, Chao-Jun Shi et al. · 0 citations
Open access Oct 2026

QASHI: Quality-Aware Scale-Protected Hybrid Inference for Small-Object Detection in Optical and SAR Remote Sensing Imagery

Object detection in high-resolution remote sensing imagery faces a resolution–context trade-off: resizing weakens small objects, whereas cropping can truncate targets and remove context. We present QASHI, a detector-retraining-free hybrid inference framework combining a full-image view, four overlapping tiles, and at m...

Ping-Peng Tang, Li-Xin He, Mo Tao et al. · 0 citations
Open access 2026

Explicit–Implicit Collaborative Attention Guided Weakly Supervised Object Detection in Remote Sensing Images

The task of detecting objects in remote sensing imagery is particularly challenging due to dramatic scale variability, cluttered backgrounds, and densely packed instances. To circumvent the prohibitive cost of instance-level annotation, weakly supervised object detection (WSOD) techniques that rely solely on image-leve...

Yue-Yi Han, Wei-Da Zhan, De-Hua Huo et al. · 0 citations
Open access Oct 2026

UniRS-Seg: Shared Sparse–Dense Query Decoding for Joint Semantic and Referring Remote Sensing Segmentation

Remote sensing semantic and referring segmentation share mask geometry but differ in whether a category vocabulary or a natural-language expression selects the output. UniRS-Seg bridges these interfaces through a shared sparse–dense query compiler and a hierarchical mask decoder, jointly trained on independently annota...

Aya Taourirte, Hai Huan · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.