岡山大学 工学部 情報工学科 （大学院自然科学研究科 計算機科学講座） 尺長 研究室

Presentation on theme: "岡山大学 工学部 情報工学科 （大学院自然科学研究科 計算機科学講座） 尺長 研究室"— Presentation transcript:

1 岡山大学 工学部 情報工学科 （大学院自然科学研究科 計算機科学講座） 尺長 研究室
最近の研究紹介 仮想光学環境 岡山大学　工学部　情報工学科 （大学院自然科学研究科　計算機科学講座） 尺長　研究室

2 実光学環境 実照明 反射特性 This slide shows an ordinary exhibition of 3d object. In the environment, a human client see a real object under a real lighting condition. 実物体 観測者

3 仮想光学環境 Projector スクリーン物体上に，様々な反射特性・仮想照明を再現 人間が持つ多様な奥行き知覚機能を自然に利用できる
反射パターン 仮想照明 Projector Projector PC The virtual photometric environment simulates such a real photometric environment by projecting a photometric pattern onto a screen object. A combination of a photometric pattern and a shape of the screen object makes a realistic exhibition of archived object. In our current prototype system, the screen object is made of white plaster, therefore the surface is regarded as a Lambertian surface. A liquid crystal projector is set up in a distance to the screen object and the surface model is made up before the projection. 観測者 スクリーン物体 スクリーン物体上に，様々な反射特性・仮想照明を再現 人間が持つ多様な奥行き知覚機能を自然に利用できる

4 Shape is acquired by active stereovision
スクリーン物体の形状計測 Shape is acquired by active stereovision Projector Camera At first, the geometric data is sampled by a conventional range finder system, which consists of a projector and a camera. After the calibration, 3d depth data is recovered pixel by pixel from a set of images of projected stripes. Screen object

5 スクリーン物体（表面）の法線 ３原色（Ｒ，Ｇ，Ｂ）で３次元直交ベクトルを表現‘
Once a 3d depth data is provided, a surface normal is estimated by principal component analysis in a local area on the surface. This slide shows color representations of surface normals of two objects, where red, green and blue color correspond to orthogonal 3d directions in 3d space. After the surface normal recovery, we can utilize a geometric surface model of the object for the further processings.

6 実サンプルによる反射特性の学習 2d BRDF map recovered from one point サンプルを装置中央に配置
光源位置を制御しながら回転 固定カメラで反射特性を計測 θi θ’r 2d BRDF map recovered from one point A simple measurement system is made up in our laboratory for learning the 2d BRDF table from a sample object. As shown in this movie, omni-directional lighting is implemented by combining the rotation control with a set of lighting switches. After a set of measurements are accomplished, a set of reflectance is plotted in the 2d table as shown in this slide. In this stage, a lot of part has no value because only a sparse sampling is accomplished in our system. （black: not sampled）

7 補間による反射特性の推定 Reconstruction of 2d BRDF Integration Interpolation
Soup bowl (lacquer ware) Sample images Reconstruction of 2d BRDF This slide shows an example of 2d BRDF reconstruction. In this example, a sample object is a soup bowl, a typical Japanese lacquer ware. After integrating the BRDF tables, we can reconstruct the BRDF table by interpolation and extrapolation. Integration Interpolation

8 形状と反射特性を用いて生成した 投影パターン
This slide shows how the BRDF table works for photometric pattern generation. Specular reflection is also appropriately simulated as well as diffuse reflection. Location of viewer = location of LCP Light direction ： moving Specular reflection is also simulated

9 仮想光学環境（プロジェクタ１台の系） This movie shows another example. In the case, a Miroku head is used as a screen object.