2:30 PM - 2:45 PM
[G03-04] Practical research on the characteristics of 3D recognition using 3D models and 3DCG
Keywords:3D recognition, 3D models, 3DCG, topography
1Introduction
In geoscience education, recognition of three-dimensional objects is an important subject. However, students often have difficulty in recognizing topographic features, even in the school environment where they usually live. Generally, topographic maps are used to recognize landforms, but it has been reported that it is difficult to read contour lines. In recent years, methods to represent topographic features in three dimensions, such as three-dimensional models and 3DCG, have been increasing and their effectiveness is being recognized.
However, how 3D models and 3DCG work on the recognition of landforms has not been examined. Understanding the characteristics of these teaching materials is useful for preparing teaching materials according to their purposes. Therefore, the purpose of this study was to investigate how 3D models and 3DCG affect students' understanding of topography.
2Research Methods
A 3D model and 3DCG of the topography were created for the area surrounding Shibuya. For the 3D model, a 3D printer was used to create a real plastic model. In creating the teaching materials, we referred to Kawashima et al. (2019) and created a plastic three-dimensional model created by a 3D printer and a transparent sheet with a topographic map on an OHP sheet.The data for creating the three-dimensional model with a 3D printer was downloaded from the H P of the Geographical Survey Institute map. The OHP sheet was printed with a topographic map of the area and the points where inundation was expected, and the scale was the same as that of the 3D model. The 3DCG of the topography was created using the method reported by Iida and Kubota (2022), which utilizes USDZ files and an iPad (Apple Inc.). In this study, 3DCGs of the relevant areas were downloaded from Geographical Survey Institute maps, and the light-colored maps were pasted onto the 3DCGs and displayed. In order to clarify the effect of understanding the topography, the map was not a flood hazard map with inundation forecasts, but a monochromatic map. In this study, the vertical scale of the target area was enlarged about 15 times to create a 3DCG. An overly large vertical magnification rate may make it difficult to understand the topographic features of the entire area. Therefore, we tried to set the magnification rate as large as possible within the range where the topographic features of the entire area can be understood. As a result, the vertical magnification was set to approximately 15 times.
In the analysis, the inundation area was set as an indicator closely related to the topography, and the participants were asked to forecast the inundation area using a three-dimensional model and a 3DCG, respectively. Nine prediction points were set, and the degree of 3D recognition of the 3D model and the 3DCG was evaluated based on the correctness of these predictions. In addition, we evaluated the differences in 3D recognition based on the students' descriptions of each material. As for the description items, we set the questions "Please tell us what you noticed, understood, or understood because of the 3D model using iPad" and "Please tell us what you noticed, understood, or understood because of the 3D model and transparent sheet" and asked the students to describe them after the class.
The class practice was targeted at third-year junior high school students in the Tokyo metropolitan area. In the class, students were first asked to make a prediction based on a topographic map, and then to observe a 3D model or a 3DCG. In the analysis, the topographic map was compared with the 3D model or 3DCG.
3Results and Discussion
The results of the inundation forecasting showed that both the 3D model and the 3DCG significantly improved the rate of correct answers to the inundation forecast compared to the topographic map. Comparing the 3D model and the 3DCG, there was no significant difference in the percentage of correct responses for rough topography, but the 3D model tended to have a higher percentage of correct responses for microtopography, such as depressions on a plateau. Furthermore, it was found that the combination of the two models resulted in more precise recognition of topographic features. In addition, quantitative text analysis was conducted on the evaluations of 3D models and 3DCG, suggesting that the points that 3DCG can zoom in and out and that 3DCG uses the sense of touch were useful for understanding topographic features. These results suggest that stereoscopic perception is different between 3D models and 3DCG, and that stereoscopic perception can be made more precise by combining the two.
This research was partially supported by the Chuden Educational Foundation.
In geoscience education, recognition of three-dimensional objects is an important subject. However, students often have difficulty in recognizing topographic features, even in the school environment where they usually live. Generally, topographic maps are used to recognize landforms, but it has been reported that it is difficult to read contour lines. In recent years, methods to represent topographic features in three dimensions, such as three-dimensional models and 3DCG, have been increasing and their effectiveness is being recognized.
However, how 3D models and 3DCG work on the recognition of landforms has not been examined. Understanding the characteristics of these teaching materials is useful for preparing teaching materials according to their purposes. Therefore, the purpose of this study was to investigate how 3D models and 3DCG affect students' understanding of topography.
2Research Methods
A 3D model and 3DCG of the topography were created for the area surrounding Shibuya. For the 3D model, a 3D printer was used to create a real plastic model. In creating the teaching materials, we referred to Kawashima et al. (2019) and created a plastic three-dimensional model created by a 3D printer and a transparent sheet with a topographic map on an OHP sheet.The data for creating the three-dimensional model with a 3D printer was downloaded from the H P of the Geographical Survey Institute map. The OHP sheet was printed with a topographic map of the area and the points where inundation was expected, and the scale was the same as that of the 3D model. The 3DCG of the topography was created using the method reported by Iida and Kubota (2022), which utilizes USDZ files and an iPad (Apple Inc.). In this study, 3DCGs of the relevant areas were downloaded from Geographical Survey Institute maps, and the light-colored maps were pasted onto the 3DCGs and displayed. In order to clarify the effect of understanding the topography, the map was not a flood hazard map with inundation forecasts, but a monochromatic map. In this study, the vertical scale of the target area was enlarged about 15 times to create a 3DCG. An overly large vertical magnification rate may make it difficult to understand the topographic features of the entire area. Therefore, we tried to set the magnification rate as large as possible within the range where the topographic features of the entire area can be understood. As a result, the vertical magnification was set to approximately 15 times.
In the analysis, the inundation area was set as an indicator closely related to the topography, and the participants were asked to forecast the inundation area using a three-dimensional model and a 3DCG, respectively. Nine prediction points were set, and the degree of 3D recognition of the 3D model and the 3DCG was evaluated based on the correctness of these predictions. In addition, we evaluated the differences in 3D recognition based on the students' descriptions of each material. As for the description items, we set the questions "Please tell us what you noticed, understood, or understood because of the 3D model using iPad" and "Please tell us what you noticed, understood, or understood because of the 3D model and transparent sheet" and asked the students to describe them after the class.
The class practice was targeted at third-year junior high school students in the Tokyo metropolitan area. In the class, students were first asked to make a prediction based on a topographic map, and then to observe a 3D model or a 3DCG. In the analysis, the topographic map was compared with the 3D model or 3DCG.
3Results and Discussion
The results of the inundation forecasting showed that both the 3D model and the 3DCG significantly improved the rate of correct answers to the inundation forecast compared to the topographic map. Comparing the 3D model and the 3DCG, there was no significant difference in the percentage of correct responses for rough topography, but the 3D model tended to have a higher percentage of correct responses for microtopography, such as depressions on a plateau. Furthermore, it was found that the combination of the two models resulted in more precise recognition of topographic features. In addition, quantitative text analysis was conducted on the evaluations of 3D models and 3DCG, suggesting that the points that 3DCG can zoom in and out and that 3DCG uses the sense of touch were useful for understanding topographic features. These results suggest that stereoscopic perception is different between 3D models and 3DCG, and that stereoscopic perception can be made more precise by combining the two.
This research was partially supported by the Chuden Educational Foundation.