# SAR Strikes Back: A New Hope for RSVQA

arXiv: https://arxiv.org/abs/2501.08131 · HTML: https://mlsanity.com/paper/2501.08131/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 50 references that are in the map.
- First posted on arXiv 14 Jan 2025.

## Builds on (most cited first, 19 of 19)

- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://mlsanity.com/paper/1908.03557/) (2019)
- [BigEarthNet: A Large-Scale Benchmark Archive For Remote Sensing Image Understanding](https://mlsanity.com/paper/1902.06148/) (2019)
- [RSVQA: Visual Question Answering for Remote Sensing Data](https://mlsanity.com/paper/2003.07333/) (2020)
- [EarthGPT: A Universal Multi-modal Large Language Model for Multi-sensor Image Comprehension in Remote Sensing Domain](https://mlsanity.com/paper/2401.16822/) (2024)
- [Multispectral Deep Neural Networks for Pedestrian Detection](https://mlsanity.com/paper/1611.02644/) (2016)
- [Deep Learning Meets SAR](https://mlsanity.com/paper/2006.10027/) (2020)
- [Remote Sensing ChatGPT: Solving Remote Sensing Tasks with ChatGPT and Visual Models](https://mlsanity.com/paper/2401.09083/) (2024)
- [SAR image despeckling through convolutional neural networks](https://mlsanity.com/paper/1704.00275/) (2017)
- [Half a Percent of Labels is Enough: Efficient Animal Detection in UAV Imagery using Deep CNNs and Active Learning](https://mlsanity.com/paper/1907.07319/) (2019)
- [SPACE: Structured Compression and Sharing of Representational Space for Continual Learning](https://mlsanity.com/paper/2001.08650/) (2020)
- [Multi-Modal Fusion Transformer for Visual Question Answering in Remote Sensing](https://mlsanity.com/paper/2210.04510/) (2022)
- [Multilingual Augmentation for Robust Visual Question Answering in Remote Sensing Images](https://mlsanity.com/paper/2304.03844/) (2023)
- [Segmentation-guided Attention for Visual Question Answering from Remote Sensing Images](https://mlsanity.com/paper/2407.08669/) (2024)
- [How to find a good image-text embedding for remote sensing visual question answering?](https://mlsanity.com/paper/2109.11848/) (2021)
- [Text-Guided Coarse-to-Fine Fusion Network for Robust Remote Sensing Visual Question Answering](https://mlsanity.com/paper/2411.15770/) (2024)
- [The curse of language biases in remote sensing VQA: the role of spatial attributes, language diversity, and the need for clear evaluation](https://mlsanity.com/paper/2311.16782/) (2023)
- [Can SAR improve RSVQA performance?](https://mlsanity.com/paper/2408.15642/) (2024)
- [A Visual Question Answering Method for SAR Ship: Breaking the Requirement for Multimodal Dataset Construction and Model Fine-Tuning](https://mlsanity.com/paper/2411.01445/) (2024)

## Built on by (most cited first, 3 of 3)

- [Checkmate: interpretable and explainable RSVQA is the endgame](https://mlsanity.com/paper/2508.13086/) (2025)
- [IC-EO: Interpretable Code-based assistant for Earth Observation](https://mlsanity.com/paper/2602.00117/) (2026)
- [Sentinel2Cap: A Human-Annotated Benchmark Dataset for Multimodal Remote Sensing Image Captioning](https://mlsanity.com/paper/2605.03189/) (2026)

## Abstract (arXiv)

Remote Sensing Visual Question Answering (RSVQA) is a task that extracts information from satellite images to answer questions in natural language, aiding image interpretation. While several methods exist for optical images with varying spectral bands and resolutions, only recently have high-resolution Synthetic Aperture Radar (SAR) images been explored. SAR's ability to operate in all weather conditions and capture electromagnetic features makes it a promising modality, yet no study has compared SAR and optical imagery in RSVQA or proposed effective fusion strategies. This work investigates how to integrate SAR data into RSVQA and how to best combine it with optical images. We present a dataset that enables SAR-based RSVQA and explore two pipelines for the task. The first is an end-to-end model, while the second is a two-stage framework: SAR information is first extracted and translated into text, which is then processed by a language model to produce the final answer. Our results show that the two-stage model performs better, improving accuracy by nearly 10% over the end-to-end approach. We also evaluate fusion strategies for combining SAR and optical data. A decision-level fusion yields the best results, with an F1-micro score of 75.00%, F1-average of 81.21%, and overall accuracy of 75.49% on the proposed dataset. SAR proves especially beneficial for questions related to specific land cover types, such as water areas, demonstrating its value as a complementary modality to optical imagery.