TY - GEN
T1 - Scene Understanding and 3D Imagination: A Comparison between Machine Learning and Human Cognition
AU - Schoosleitner, Michael
AU - Ullrich, Torsten
PY - 2020
Y1 - 2020
N2 - Spatial perception and three-dimensional imagination are important characteristics for many construction tasks in civil engineering. In order to support people in these tasks, worldwide research is being carried out on assistance systems based on machine learning and augmented reality. In this paper, we examine the machine learning component and compare it to human performance. The test scenario is to recognize a partly-assembled model, identify its current status, i.e. the current instruction step, and to return the next step. Thus, we created a database of 2D images containing the complete set of instruction steps of the corresponding 3D model. Afterwards, we trained the deep neural network RotationNet with these images. Usually, the machine learning approaches are compared to each other; our contribution evaluates the machine learning results with human performance tested in a survey: in a clean-room setting the survey and RotationNet results are comparable and neither is significa ntly better. The real-world results show that the machine learning approaches need further improvements
AB - Spatial perception and three-dimensional imagination are important characteristics for many construction tasks in civil engineering. In order to support people in these tasks, worldwide research is being carried out on assistance systems based on machine learning and augmented reality. In this paper, we examine the machine learning component and compare it to human performance. The test scenario is to recognize a partly-assembled model, identify its current status, i.e. the current instruction step, and to return the next step. Thus, we created a database of 2D images containing the complete set of instruction steps of the corresponding 3D model. Afterwards, we trained the deep neural network RotationNet with these images. Usually, the machine learning approaches are compared to each other; our contribution evaluates the machine learning results with human performance tested in a survey: in a clean-room setting the survey and RotationNet results are comparable and neither is significa ntly better. The real-world results show that the machine learning approaches need further improvements
UR - http://www.scopus.com/inward/record.url?scp=85083516330&partnerID=8YFLogxK
U2 - 10.5220/0009350002310238
DO - 10.5220/0009350002310238
M3 - Conference paper
VL - 2, HUCAPP
T3 - VISIGRAPP 2020 - Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
SP - 231
EP - 238
BT - Proceedings of the International Joint Conference on Computer Vision and Computer Graphics Theory and Applications
A2 - Chessa, Manuela
A2 - Paljic, Alexis
A2 - Braz, Jose
PB - SciTePress
T2 - 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Y2 - 8 February 2021 through 10 February 2021
ER -