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

arXiv: https://arxiv.org/abs/2609.25584 · HTML: https://mlsanity.com/paper/2609.25584/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 38 references that are in the map.
- First posted on arXiv 22 Sep 2026.

## Builds on (most cited first, 22 of 22)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [YOLO9000: Better, Faster, Stronger](https://mlsanity.com/paper/1612.08242/) (2016)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [Open-vocabulary Object Detection via Vision and Language Knowledge Distillation](https://mlsanity.com/paper/2104.13921/) (2021)
- [MDETR -- Modulated Detection for End-to-End Multi-Modal Understanding](https://mlsanity.com/paper/2104.12763/) (2021)
- [Detecting Twenty-thousand Classes using Image-level Supervision](https://mlsanity.com/paper/2201.02605/) (2022)
- [Object Detection in Optical Remote Sensing Images: A Survey and A New Benchmark](https://mlsanity.com/paper/1909.00133/) (2019)
- [Equalization Loss for Long-Tailed Object Recognition](https://mlsanity.com/paper/2003.05176/) (2020)
- [Object Detection in Aerial Images: A Large-Scale Benchmark and Challenges](https://mlsanity.com/paper/2102.12219/) (2021)
- [DetCLIP: Dictionary-Enriched Visual-Concept Paralleled Pre-training for Open-world Detection](https://mlsanity.com/paper/2209.09407/) (2022)
- [xView: Objects in Context in Overhead Imagery](https://mlsanity.com/paper/1802.07856/) (2018)
- [Seesaw Loss for Long-Tailed Instance Segmentation](https://mlsanity.com/paper/2008.10032/) (2020)
- [Simple multi-dataset detection](https://mlsanity.com/paper/2102.13086/) (2021)
- [Towards Large-Scale Small Object Detection: Survey and Benchmarks](https://mlsanity.com/paper/2207.14096/) (2022)
- [T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy](https://mlsanity.com/paper/2403.14610/) (2024)
- [Object Detection with a Unified Label Space from Multiple Datasets](https://mlsanity.com/paper/2008.06614/) (2020)
- [Locate Anything on Earth: Advancing Open-Vocabulary Object Detection for Remote Sensing Community](https://mlsanity.com/paper/2408.09110/) (2024)
- [Detection Hub: Unifying Object Detection Datasets via Query Adaptation on Language Embedding](https://mlsanity.com/paper/2206.03484/) (2022)
- [WeDetect: Fast Open-Vocabulary Object Detection as Retrieval](https://mlsanity.com/paper/2512.12309/) (2025)
- [Open-Vocabulary Object Detection via Language Hierarchy](https://mlsanity.com/paper/2410.20371/) (2024)

## Built on by (most cited first, 1 of 1)

- [SetOPD: From Few Visual Exemplars to Multimodal Candidate Sets for Remote-Sensing Open-Prompt Detection](https://mlsanity.com/paper/2609.32529/) (2026)

## Abstract (arXiv)

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 textitcar and textitvan under atomic prompts yet miss the same instances under textitvehicle; 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.