# Underwater Object Classification and Detection: first results and open challenges

arXiv: https://arxiv.org/abs/2201.00977 · HTML: https://mlsanity.com/paper/2201.00977/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 30 references that are in the map.
- First posted on arXiv 4 Jan 2022.

## Builds on (most cited first, 17 of 17)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [SSD: Single Shot MultiBox Detector](https://mlsanity.com/paper/1512.02325/) (2015)
- [Fast R-CNN](https://mlsanity.com/paper/1504.08083/) (2015)
- [Rich feature hierarchies for accurate object detection and semantic segmentation](https://mlsanity.com/paper/1311.2524/) (2013)
- [Rethinking ImageNet Pre-training](https://mlsanity.com/paper/1811.08883/) (2018)
- [Fast Underwater Image Enhancement for Improved Visual Perception](https://mlsanity.com/paper/1903.09766/) (2019)
- [Enhancing Underwater Imagery using Generative Adversarial Networks](https://mlsanity.com/paper/1801.04011/) (2018)
- [Robust Gesture-Based Communication for Underwater Human-Robot Interaction in the context of Search and Rescue Diver Missions](https://mlsanity.com/paper/1810.07122/) (2018)
- [WQT and DG-YOLO: towards domain generalization in underwater object detection](https://mlsanity.com/paper/2004.06333/) (2020)
- [Reveal of Domain Effect: How Visual Restoration Contributes to Object Detection in Aquatic Scenes](https://mlsanity.com/paper/2003.01913/) (2020)
- [Towards a Generic Diver-Following Algorithm: Balancing Robustness and Efficiency in Deep Visual Detection](https://mlsanity.com/paper/1809.06849/) (2018)
- [Hypothesis-based Belief Planning for Dexterous Grasping](https://mlsanity.com/paper/1903.05517/) (2019)
- [Metrics and Benchmarks for Remote Shared Controllers in Industrial Applications](https://mlsanity.com/paper/1906.08381/) (2019)
- [Automatic Detection of Myocontrol Failures Based upon Situational Context Information](https://mlsanity.com/paper/1906.11564/) (2019)

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

- [SyreaNet: A Physically Guided Underwater Image Enhancement Framework Integrating Synthetic and Real Images](https://mlsanity.com/paper/2302.08269/) (2023)
- [A Real-Time Framework for Domain-Adaptive Underwater Object Detection with Image Enhancement](https://mlsanity.com/paper/2403.19079/) (2024)
- [Robot Grasping and Manipulation: A Prospective](https://mlsanity.com/paper/2303.07807/) (2023)
- [A biological vision inspired framework for machine perception of abutting grating illusory contours](https://mlsanity.com/paper/2508.17254/) (2025)

## Abstract (arXiv)

This work reviews the problem of object detection in underwater environments. We analyse and quantify the shortcomings of conventional state-of-the-art (SOTA) algorithms in the computer vision community when applied to this challenging environment, as well as providing insights and general guidelines for future research efforts. First, we assessed if pretraining with the conventional ImageNet is beneficial when the object detector needs to be applied to environments that may be characterised by a different feature distribution. We then investigate whether two-stage detectors yields to better performance with respect to single-stage detectors, in terms of accuracy, intersection of union (IoU), floating operation per second (FLOPS), and inference time. Finally, we assessed the generalisation capability of each model to a lower quality dataset to simulate performance on a real scenario, in which harsher conditions ought to be expected. Our experimental results provide evidence that underwater object detection requires searching for "ad-hoc" architectures than merely training SOTA architectures on new data, and that pretraining is not beneficial.