the puzzle has remained unsolved for decades, and now, a team led by the ca' foscari university of venicesystem to analyse and eventually piece together the frescoes. called repair, or reconstructing the past: artificial intelligence and robotics meet cultural heritage, it's the first time machine learning and artificial intelligence techniques will be used to take on a project of this scale. the robot is scanning a piece of fresco using polarised lenses and so, at the end of the process, the piece will be scanned in 3d. with the same infrastructure and the same type of technology, we can also scan the same piece using hyperspectral sensors that are able to collect information that the human eyes cannot see. this information includes the residual colours of pigments used by the romans which cannot be seen with the naked eye. what we can see here is basically, the light coming from our object at different wavelengths. we can select the wavelength that we want to look at, and this amount gave us initial information of our material of oursample. different points on the objects can be selecte

the puzzle has remained unsolved for decades, and now, a team led by ca' foscari university of venicelled repair, or reconstructing the past: artificial intelligence and robotics meet cultural heritage, it's the first time machine learning and artificial intelligence techniques will be used to take on a project of this scale. the robot is scanning a piece of fresco using polarised lenses and so, at the end of the process, the piece will be scanned in 3d. with the same infrastructure and the same type of technology, we can also scan the same piece using hyperspectral sensors that are able to collect information that the human eyes cannot see. this information includes the residual colours of pigments used by the romans which cannot be seen with the naked eye. what we can see here is basically, the light coming from our object at different wavelengths. we can select the wavelength that we want to look at, and this amount gave us initial information of our material of our sample. different points on the objects can be selected to examine closely, with information given about similarities

the puzzle has remained unsolved for decades, and now, a team led by the ca�* foscari university of veniceusing polarised lenses and so, at the end of the process, the piece will be scanned in 3d. with the same infrastructure and the same type of technology, we can also scan the same piece using hyperspectral sensors that are able to collect information that the human eyes cannot see. this information includes the residual colours of pigments used by the romans which cannot be seen with the naked eye. what we can see here is basically, the light coming from our object at different wavelengths. we can select the wavelength that we want to look at, and this amount gave us initial information of our material of our sample. different points on the objects can be selected to examine closely, with information given about similarities and differences between pixels in the image. once all of the information is collected, an algorithm will suggest how it thinks the pieces fit together and what�*s missing, and run it past a human expert. if this works, i think it will have a huge potential in future

the puzzle has remained unsolved for decades, and now, a team led by the ca' foscari university of veniceot is scanning a piece of fresco using polarised lenses and so, at the end of the process, the piece will be scanned in 3d. with the same infrastructure and the same type of technology, we can also scan the same piece using hyperspectral sensors that are able to collect information that the human eyes cannot see. this information includes the residual colours of pigments used by the romans which cannot be seen with the naked eye. what we can see here is basically, the light coming from our object at different wavelengths. we can select the wavelength that we want to look at, and this amount gave us initial information of our material of our sample. different points on the objects can be selected to examine closely, with information given about similarities and differences between pixels in the image. once all of the information is collected, an algorithm will suggest how it thinks the pieces fit together and what's missing, and run it past a human expert. if this works, i think it will