From: Ander A. <aar...@vi...> - 2015-08-27 12:01:33
|
Dear Yeonsoo, dear Toshiba co-workers, Firstly, I must congratulate you and your Toshiba co-workers for your good work and your contributions to the volume rendering component in X3DOM. Regarding your question, from the three options you are proposing, I would prefer the second one (2). The ProjectionVolumeStyle node has a quite specific use case in the medical visualization field (e.g, the MaximunIntensityProjection), >From the X3D standard point of view and noticing that your needs are related to the creation of a custom style to visualize radar data, I have some thoughts for a future integration of your proposed DepthMappedVolumeStyle into the other volume nodes already defined by the X3D spec. 1. From my point of view, the depth texture should be part of the X3DVolumeDataNode node, required when intersections of 3D objects are to be expected. This behaviour encloses your radar visualization use case regarding the intersection of the radar volume and the surrounding GIS data. 2. The isoCutOff (minimum and maximum values) could also be achieved with the modification of the alpha channel of the transfer function (OpacityMapVolumeStyle). By dynamically changing the values (user interface), the changes would be directly applied to the GLSL shaders via the canvas texture. 3. Regarding the xSectionPosition and xSectionOrientation attributes, I noticed that they are streamlined directly to the GLSL shaders. This clipping behaviour is therefore only usable in this proposed node. From my point of view, I see two ways to extend this behaviour to the X3D specs in a more generic way: - Using already existing ClippingPlane node entity. One or more clippings planes defined in the X3D scene would be considered in the volume rendering algorithms to do the clipping of the volume. In this case, those planes would clip the volume and any other 3D objects in the scene. For example, imagine that there is a VolumeData in this demo ( http://examples.x3dom.org/clipPlane/clipplane.html). Any added clipping plane would clip the 3D geometry and the volume data. - Adding a new composable cross-sectional style to the X3D standard. In this case, all composable styles will benefit for a multi cross-sectional view, but they would be only applied to the volume data node, without affecting the rest of the 3D scene. I imagine this behaviour would fit better you radar volume visualization application, since you don't need to clip the GIS data, just the volume data. Summarizing, the idea behind the proposed new node is very good and consistent with my ideas for the improvement of the X3D standard itself, and the X3DOM framework. But I would use this opportunity to discuss a consistent proposal that could end in the X3D standard following the official mechanisms (if I remember correctly, the Volume Rendering nodes are spec'd by the X3D Medical Working Group). Any thoughts? -- Ander Arbelaiz Aranzasti Ayudante de Investigación / Research Assistant Industria y Fabricación Avanzada / Industry and Advanced Manufacturing Vicomtech-IK4 - Visual Interaction Communication Technologies Mikeletegi Pasealekua, 57 - Parque Tecnológico 20009 Donostia - San Sebastián - Spain Tel: +[34] 943 30 92 30 Fax: +[34] 943 30 93 93 e-mail: aar...@vi... *** member of IK4 Research Alliance ****www.ik4.es *** member of GraphicsMedia.net ****www.graphicsmedia.net ----------------------------------------------------- Vicomtech-IK4 is an ISO 9001:2000 certified institute ----------------------------------------------------- Aviso Legal - Política de privacidad (http://www.vicomtech.org/proteccion-datos) Lege Oharra - Pribatutasun politika (http://www.vicomtech.org/eu/proteccion-datos) Legal Notice - Privacy policy (http://www.vicomtech.org/en/proteccion-datos) Dear Fraunhofer members (Johannes Behr, Max Limper, Tobias Alexander > Franke), Vicomtech members, and someone who are so much interested in > X3D/X3DOM Volume Rendering Component > > > As I mentioned during the paper session presentation at Web3D 2015 Crete, > we will commit RadarVolumeStyle, changing the name as > DepthMappedVolumeStyle node to X3DOM 1.7. > > > But if we add the depthTexture for polygons, transferFunction fields and > xSectionPosition, xSectionOrientation for cutting plane to > ProjectionVolumeStyle then we don?t need DepthMappedVolumeStyle. > Or add cutting plane field (xSectionPosition, xSectionOrientation) > ProjectionVolumeStyle and keep DepthMappedVolumeStyle. > Do you have any preference among those options ? > > > To wrap up options, > (1) Extend ProjectionVolumeStyle (fields: depthTexture, > transferFunction fields and xSectionPosition, xSectionOrientation) > (2) Add DepthMappedVolumeStyle (fields: depthTexture, transferFunction > fields and xSectionPosition, xSectionOrientation) -> current > implementation and it will be pushed to GitHub. > (3) Add DepthMappedVolumeStyle (fields: xSectionPosition, > xSectionOrientation) and extend ProjectionVolumeStyle (xSectionPosition, > xSectionOrientation) > > > > > * Reference: https://vimeo.com/103145827 > > > Best regards, > > > Yeonsoo > > > ----------------------------------- > Yeonsoo Yang > > > Advanced Software Technology Dept. > IoT Technology Center > Industrial ICT Solutions Company > Toshiba Corporation > 1, Komukai-Toshiba-cho, Saiwai-ku, > Kawasaki, 212-8582, Japan > Tel +81-44-549-2457 > Fax +81-44-549-2443 > E-mail: yeo...@to...<mailto:yeo...@to...> > ----------------------------------- |