# Few-Shot Open-Vocabulary Remote Sensing Segmentation via Textual Inversion

arXiv: https://arxiv.org/abs/2607.25563 · HTML: https://mlsanity.com/paper/2607.25563/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 34 references that are in the map.
- First posted on arXiv 28 Jul 2026.

## Builds on (most cited first, 28 of 28)

- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [Learning to Prompt for Vision-Language Models](https://mlsanity.com/paper/2109.01134/) (2021)
- [An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion](https://mlsanity.com/paper/2208.01618/) (2022)
- [Conditional Prompt Learning for Vision-Language Models](https://mlsanity.com/paper/2203.05557/) (2022)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [Language-driven Semantic Segmentation](https://mlsanity.com/paper/2201.03546/) (2022)
- [Open-Vocabulary Semantic Segmentation with Mask-adapted CLIP](https://mlsanity.com/paper/2210.04150/) (2022)
- [Mind the Gap: Understanding the Modality Gap in Multi-modal Contrastive Representation Learning](https://mlsanity.com/paper/2203.02053/) (2022)
- [PANet: Few-Shot Image Semantic Segmentation with Prototype Alignment](https://mlsanity.com/paper/1908.06391/) (2019)
- [RemoteCLIP: A Vision Language Foundation Model for Remote Sensing](https://mlsanity.com/paper/2306.11029/) (2023)
- [Images Speak in Images: A Generalist Painter for In-Context Visual Learning](https://mlsanity.com/paper/2212.02499/) (2022)
- [LoveDA: A Remote Sensing Land-Cover Dataset for Domain Adaptive Semantic Segmentation](https://mlsanity.com/paper/2110.08733/) (2021)
- [Personalize Segment Anything Model with One Shot](https://mlsanity.com/paper/2305.03048/) (2023)
- [CAT-Seg: Cost Aggregation for Open-Vocabulary Semantic Segmentation](https://mlsanity.com/paper/2303.11797/) (2023)
- [iSAID: A Large-scale Dataset for Instance Segmentation in Aerial Images](https://mlsanity.com/paper/1905.12886/) (2019)
- [Hypercorrelation Squeeze for Few-Shot Segmentation](https://mlsanity.com/paper/2104.01538/) (2021)
- [Matcher: Segment Anything with One Shot Using All-Purpose Feature Matching](https://mlsanity.com/paper/2305.13310/) (2023)
- [OpenEarthMap: A Benchmark Dataset for Global High-Resolution Land Cover Mapping](https://mlsanity.com/paper/2210.10732/) (2022)
- [AD-CLIP: Adapting Domains in Prompt Space Using CLIP](https://mlsanity.com/paper/2308.05659/) (2023)
- [VDD: Varied Drone Dataset for Semantic Segmentation](https://mlsanity.com/paper/2305.13608/) (2023)
- [SegEarth-OV3: Exploring SAM 3 for Open-Vocabulary Semantic Segmentation in Remote Sensing Images](https://mlsanity.com/paper/2512.08730/) (2025)
- [OpenDAS: Open-Vocabulary Domain Adaptation for 2D and 3D Segmentation](https://mlsanity.com/paper/2405.20141/) (2024)
- [Tackling Few-Shot Segmentation in Remote Sensing via Inpainting Diffusion Model](https://mlsanity.com/paper/2503.03785/) (2025)
- [Textual Inversion for Efficient Adaptation of Open-Vocabulary Object Detectors Without Forgetting](https://mlsanity.com/paper/2508.05323/) (2025)
- [SeMoBridge: Semantic Modality Bridge for Efficient Few-Shot Adaptation of CLIP](https://mlsanity.com/paper/2509.26036/) (2025)
- [Few-Shot Semantic Segmentation Meets SAM3](https://mlsanity.com/paper/2604.05433/) (2026)
- [SegRAG: Training-Free Retrieval-Augmented Semantic Segmentation](https://mlsanity.com/paper/2605.17630/) (2026)

## Abstract (arXiv)

Open-vocabulary segmentation labels arbitrary categories from a text query without per-class training, yet on remote sensing imagery it underperforms on categories it handles reliably elsewhere. We find that much of this gap traces to the text query rather than to the segmentation model. Because these models are not specialized for overhead imagery, the class name that serves as the query is often a weak address into the vision-language embedding space. We show that a better name repairs part of the gap, while the remaining failures call for an address that the tested natural-language rephrasings do not provide. We recover that address from a few examples through textual inversion on a frozen model, keeping inference text only. On a representative benchmark this raises the mean intersection over union on the affected categories from 3.9 to 39.4, and across eight remote sensing datasets it improves over few-shot methods that instead inject visual prompts at inference.