Papers for

autonomous underwater vehicle developers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Underwater vision language model improves spatial reasoning with geometry data

USI-Brain: Enhancing Underwater Spatial Intelligence via Spatial Data-Driven Vision-Language Modeling

Abstract: Underwater spatial reasoning depends on object boundaries and geometric cues that are obscured by scattering, color distortion, and low visibility. Existing underwater resources capture these cues through images, annotations, and depth maps, but provide fragmented supervision for vision-language models. We present a spatial data engine that connects visual appearance with object-level geometry through enhancement, 2D and geometry-assisted question generation, and expert filtering. It produces 4.27M QA pairs across five capability groups. We use this corpus to train USI-Brain and introduce an Underwater Prior-guided Rectifier (UPR) that applies gated corrections to visual features. USI-Bench evaluates the resulting capabilities with 6,525 held-out QA pairs. On its 15 task metrics, USI-Brain and USI-Brain + UPR improve over Qwen3-VL-8B-Instruct by 5.76 and 7.63 points on average, respectively, using signed differences on 100-point scales. Ablations show that geometry-assisted supervision improves depth and ordering by 4.8 and 6.1 points with UPR disabled, while UPR adds 18.0 points in scene description under matched training data. Both variants use raw images at evaluation. Across NautData, UWBench, and MarineEval, USI-Brain achieves the best or joint-best score on 28 of 32 metrics. Model weights and USI-Bench will be released at https://github.com/SKY-Zhu-91/USI-Brain.

Wed 4 NovComputer Vision and Pattern Recognition
The gist
Underwater images are hard to understand because water blurs and distorts shapes and colors, making it tough to tell where objects really are. The authors created a large set of questions and answers that connect what things look like in underwater photos with their 3D shapes to help computers learn better. They trained a model called USI-Brain that uses this information to understand underwater scenes more accurately, especially after adding a special correction step. This model scored better than previous ones on tests measuring how well it interprets underwater spaces and objects.
Open → 2611.06040v1

Image quality alone does not predict sonar detection success

Beyond Pixel Similarity: Task-Aware Evaluation of GAN-Based Synthetic Sonar Data for Robotic Perception

Abstract: Synthetic data can reduce the cost of collecting and annotating training data for robotic perception, but generating sensor observations that preserve the characteristics relevant to downstream perception remains challenging, particularly for sonar imagery. In this work, we investigate whether conventional image-fidelity metrics adequately reflect the downstream perception performance of GAN-generated synthetic sonar data. We employ a Pix2Pix conditional generative adversarial network with four discriminator configurations characterized by different receptive fields: PixelGAN, PatchGAN-16, PatchGAN-70, and ImageGAN. The models are trained using sonar imagery from two datasets and evaluated using conventional image-fidelity metrics, including Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR), and Mean Squared Error (MSE). To complement these pixel-level measures with task-oriented evaluation, YOLOX-S, YOLOX-L, and Faster R-CNN detectors are trained exclusively on real sonar imagery and subsequently evaluated on the GAN-generated images using identical test samples and annotations across all discriminator configurations. The results reveal a discrepancy between image-fidelity and downstream object-detection performance: the configuration achieving the best SSIM, PSNR, and MSE does not consistently yield the best detection performance. In particular, PatchGAN configurations achieve strong downstream detection results despite not achieving the highest pixel-level similarity scores. These findings suggest, for the datasets and models considered, pixel-level image-fidelity metrics alone may not consistently capture the task-relevant realism of synthetic sonar observations and motivate the use of task-aware evaluation for synthetic sensor data intended for robotic perception.

Wed 16 SeptRoboticsComputer Vision and Pattern RecognitionMachine Learning
The gist
Generating fake sonar images for training robots can save time and money, but making these images look real doesn’t always make them useful for spotting objects. The authors show that common measures of image quality don’t always match how well object detectors work on those images. They tested different ways of making fake sonar pictures and found some that help detection even though they don’t score highest on image quality tests. This means it’s important to check synthetic images by how well they help with tasks, not just by how good they look.
Open → 2609.18100v1

Search based testing finds vision language model errors in underwater robot software

Search-Based Metamorphic Testing of Vision-Language Models in Autonomous Underwater Robotic Software

Abstract: Our industry partner focuses on quality assurance for industrial systems across multiple domains, including maritime systems, such as overwater vessels and autonomous underwater robots (AURs). Despite the strong performance of vision-language models (VLMs) in scene understanding, image captioning, and object recognition, their use in AUR software operating in underwater environments is underexplored. Therefore, in this context, it is important to evaluate the quality of VLMs for integration into AUR software and, so, automated software testing tools are needed to assess their suitability and improve their dependability. To this end, we propose a search-based metamorphic testing approach (MetaVLM) that identifies a minimal set of transformations on underwater images to induce incorrect model predictions, thereby revealing VLM failures. We employ NSGA-II as a multi-objective search algorithm and evaluate it over open-source VLMs, BLIP and CLIP, against a random search baseline. Results demonstrate the strengths and limitations of each VLM in the context of AUR software systems. Based on the results, we derive lessons for software engineering practitioners and researchers working on quality assurance of VLM-based software systems.

Tue 15 SeptSoftware Engineering
The gist
Underwater robots rely on vision-language models (VLMs) to understand scenes and identify objects, but these models have not been well tested underwater. The authors developed a method called MetaVLM that automatically changes underwater images in small ways to find when these models make mistakes. They used an advanced search technique to efficiently find these image changes and tested two popular models, BLIP and CLIP. Their work shows where these models perform well or fail when used in underwater robot software and offers advice for developers working with such AI systems.
Open → 2609.17007v1