Tranfer Learning of Semantic Segmentation Methods for Identifying Buried Archaeological Structures on Lidar Data

IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp. 6987–6990

Abstract

When applying deep learning to remote sensing data in archaeological research, a notable obstacle is the limited availability of suitable datasets for training models. The application of transfer learning is frequently employed to mitigate this drawback. However, there is still a need to explore its effectiveness when applied across different archaeological datasets. This paper compares the performance of various transfer learning configurations using two semantic segmentation deep neural networks on two LiDAR datasets. The experimental results indicate that transfer learning-based approaches in archaeology can lead to performance improvements, although a systematic enhancement has not yet been observed. We provide specific insights about the validity of such techniques that can serve as a baseline for future works.

Authors 6

1. Gregory Sech

   Italian Institute of Technology · Center for Cultural Heritage Technology

   Affiliation as printed

   Istituto Italiano di Tecnologia,Center for Cultural Heritage Technology,Venice,Italy

   Center for Cultural Heritage Technology, Istituto Italiano di Tecnologia, Venice, Italy

2. Paolo Soleni

   Italian Institute of Technology · Center for Cultural Heritage Technology

   Affiliation as printed

   Istituto Italiano di Tecnologia,Center for Cultural Heritage Technology,Venice,Italy

   Center for Cultural Heritage Technology, Istituto Italiano di Tecnologia, Venice, Italy

3. Wouter Verschoof‐van der Vaart Aachen Faculty of Archaeology, Leiden Faculty of Archaeology, Leiden University

   Leiden University

   Affiliation as printed

   Leiden University,Faculty of Archaeology,Leiden,Netherlands

   Faculty of Archaeology, Leiden University, Leiden, Netherlands

4. Žiga Kokalj

   Slovenian Academy of Sciences and Arts · Research Centre of the Slovenian Academy of Sciences and Arts

   Affiliation as printed

   Research Centre of the Slovenian Academy of Sciences and Arts,Ljubljana,Slovenia

   Research Centre of the Slovenian Academy of Sciences and Arts, Ljubljana, Slovenia

5. Arianna Traviglia

   Italian Institute of Technology · Center for Cultural Heritage Technology

   Affiliation as printed

   Istituto Italiano di Tecnologia,Center for Cultural Heritage Technology,Venice,Italy

   Center for Cultural Heritage Technology, Istituto Italiano di Tecnologia, Venice, Italy

6. Marco Fiorucci

   Kyoto University · Italian Institute of Technology · Center for Cultural Heritage Technology

   Affiliation as printed

   Istituto Italiano di Tecnologia,Center for Cultural Heritage Technology,Venice,Italy

   Center for Cultural Heritage Technology, Istituto Italiano di Tecnologia, Venice, Italy

   Kyoto University, Kyoto, Japan