Related Experiment Video
Updated: May 8, 2026

06:54
Photorealistic Learned Landscapes for Augmented Reality
Published on: June 27, 2025
A comparative analysis between active and passive techniques for underwater 3D reconstruction of close-range objects.
Gianfranco Bianco1, Alessandro Gallo, Fabio Bruno
1Department of Mechanical Engineering, University of Calabria, Cosenza 87036, Italy. gianfranco.bianco@unical.it
Sensors (Basel, Switzerland)
|August 23, 2013
Summary
This study compares active and passive 3D imaging techniques for underwater data collection in turbid conditions. Results offer insights into 3D point acquisition capabilities for marine research and archaeology.
Area of Science:
- Marine Biology
- Underwater Archaeology
- 3D Imaging Technologies
Background:
- Collecting 3D data underwater is crucial for fields like marine biology and archaeology.
- Existing 3D imaging techniques are often used for close-range underwater acquisitions.
Purpose of the Study:
- To compare the performance of active and passive 3D imaging techniques in underwater environments.
- To evaluate their capability in acquiring 3D points under varying turbidity conditions.
Main Methods:
- Two whole-field, stereo-configured 3D imaging techniques (active and passive) were compared.
- A laboratory setup simulated poor visibility using clay suspension in a water tank.
- Acquisition parameters like setup, working distance, and calibration were kept consistent.
Main Results:
- The study evaluated the accuracy and density of 3D points acquired by both techniques.
- Preliminary analysis focused on their performance in turbid water conditions.
- Results provide a baseline for comparing 3D techniques in challenging aquatic environments.
Conclusions:
- The comparative analysis provides foundational data on active versus passive 3D imaging in turbid water.
- While not yet for real-world application, the findings guide future research in underwater 3D data acquisition.
- This research serves as a reference for evaluating other 3D techniques and algorithms in similar conditions.

