# Pre-trained Models for Sonar Images

arXiv: https://arxiv.org/abs/2108.01111 · HTML: https://mlsanity.com/paper/2108.01111/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 16 references that are in the map.
- First posted on arXiv 2 Aug 2021.

## Builds on (most cited first, 11 of 11)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size](https://mlsanity.com/paper/1602.07360/) (2016)
- [The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale](https://mlsanity.com/paper/1811.00982/) (2018)
- [CNN Features off-the-shelf: an Astounding Baseline for Recognition](https://mlsanity.com/paper/1403.6382/) (2014)
- [What makes ImageNet good for transfer learning?](https://mlsanity.com/paper/1608.08614/) (2016)
- [Real-time Convolutional Neural Networks for Emotion and Gender Classification](https://mlsanity.com/paper/1710.07557/) (2017)
- [Deep Neural Networks for Marine Debris Detection in Sonar Images](https://mlsanity.com/paper/1905.05241/) (2019)
- [Best Practices in Convolutional Networks for Forward-Looking Sonar Image Recognition](https://mlsanity.com/paper/1709.02601/) (2017)

## Built on by (most cited first, 4 of 4)

- [Self-supervised Learning for Sonar Image Classification](https://mlsanity.com/paper/2204.09323/) (2022)
- [The Marine Debris Forward-Looking Sonar Datasets](https://mlsanity.com/paper/2503.22880/) (2025)
- [Weakly Supervised Object Segmentation by Background Conditional Divergence](https://mlsanity.com/paper/2506.22505/) (2025)
- [BenthicDINO: Physics-Informed Self-Distillation for View-Invariant Side-Scan Sonar Representations](https://mlsanity.com/paper/2608.23215/) (2026)

## Abstract (arXiv)

Machine learning and neural networks are now ubiquitous in sonar perception, but it lags behind the computer vision field due to the lack of data and pre-trained models specifically for sonar images. In this paper we present the Marine Debris Turntable dataset and produce pre-trained neural networks trained on this dataset, meant to fill the gap of missing pre-trained models for sonar images. We train Resnet 20, MobileNets, DenseNet121, SqueezeNet, MiniXception, and an Autoencoder, over several input image sizes, from 32 x 32 to 96 x 96, on the Marine Debris turntable dataset. We evaluate these models using transfer learning for low-shot classification in the Marine Debris Watertank and another dataset captured using a Gemini 720i sonar. Our results show that in both datasets the pre-trained models produce good features that allow good classification accuracy with low samples (10-30 samples per class). The Gemini dataset validates that the features transfer to other kinds of sonar sensors. We expect that the community benefits from the public release of our pre-trained models and the turntable dataset.