Mapping the seabed with cutting-edge research and applications in photogrammetric computer vision, image analysis and remote sensing

About us

3[Deep]Vision is an Award Winning team performing cutting-edge research and developing applications in the intersection of machine learning, computer vision and earth observation of the optically shallow seabed.

The determination of accurate bathymetric information is a key element for near offshore activities, hydrological studies such as coastal engineering applications, sedimentary processes, hydrographic surveying as well as archaeological mapping and biological research. UAV imagery processed with Structure from Motion (SfM) and Multi View Stereo (MVS) techniques or Satellite imagery processed with empirical or deep learning methods, can provide a low-cost alternative to established shallow seabed mapping techniques offering the important visual information as well. 3[Deep]Vision group, in order to overcome the water refraction errors, employ machine learning tools that are able to learn the systematic underestimation of the estimated depths and deliver bathymetric information in a centimeter accuracy level.

Research Applications: Shallow water mapping for Constructions, Underwater Archaeology, Biological Monitoring, Benthic Cover Mapping, Marine Litter Detection

We use UAV-borne, airborne or satellite-borne imagery to generate seamless orthoimages and maps of the land and the seabed with very accurate depth contours.

Accurate and very detailed bathymetric information in the form of 3D point clouds, 3D meshes or colourscaled images

Semantic classification of the seabed.

We have big experience in seabed mapping applications using Earth Observation data and Machine Learning, having developed robust and extremely reliable models to compensate for refraction effects or predict bathymetry.

Awarded by the EU Agency for the Space Programme (EUSPA)

We have been awared with the 1st Prize in the GeomaticsOnTheMove contest 2020!

Read more
The highest bathymetric accuracy available

3 x more accurate results compared to the rest of the refraction correction methods worldwide

Read more
Refraction Correction

Refraction-free image-based 3D point clouds with colour information and images with very high accuracy - the highest available in the market and academic research!

Read more
Semantinc Information

Semantically enriched products using Machine Learning and Deep Learning based methods for semantic classification and segmentation of 3D point clouds and images!


Projects for seabed mapping




Panagiotis Agrafiotis


Alexandra Papadaki


Dimitrios Skarlatos


Konstantinos Karantzalos


Contact Us

Your message has been sent. Thank you!