I added a couple of images into a GroupLayer. Does anyone know how to create an image from that GroupLayer?
Thanks.
Just a quick idea:
Surface.drawLayer(Layer layer);
Renders the supplied scene graph into this surface. This renders the scene graph into the surface's texture using the main render pipeline, so custom shaders (on platforms that support them) are handled properly. NOTE: if there is a loop in the scene graph (like you add an ImmediateLayer that calls surf.drawLayer(graphics().rootLayer())) you will get the infinite loop that you deserve. Caveat renderer.
Related
I need to pass to a native method the handle of a javaFX window on which to render a world made with OSG (Open Scene Graph).
I understood that on javaFX only the Stage has this native handle, all the other objects, for example Nodes and Pane do not have a native handle.
So I thought, if it were possible to insert the Stage (for native rendering) and javafx widgets into the primary Stage, I would have the possibility to manage different levels, so that I could superimpose javaFX objects on the OSG rendering.
Do you have any idea?
I have a requirement to build multiple scenes in web VR A-frame. Each scene will have a button which when clicked, will load a new scene with different background.
What is the recommended way to build scenes in A-frame?
Create just one scene and swap the entities at runtime. Each entity can correspond to a button, background, etc.
Create multiple pages(eg. index.html), each having its own scene and load each page on click event.
Approach (1) seems to be the preferred one as the second approach would mean that 'Back' button in browser will be enabled on loading new pages, which is undesirable and affects the user experience in VR.
Can anyone confirm that approach (1) is preferred?
Definitely a better experience if you can keep everything in a "single-page app" (1) and swap out entities. Especially if you are doing something as simple as swapping a background. Only a couple browsers have implemented proper in-VR link traversal, and the link traversal UX at the moment is non-existent.
https://aframe.io/docs/0.6.0/guides/building-a-360-image-gallery.html
Using Cocos3D, is it possible to take screenshot of the 3D model in the background without the user knowing it?
For pre-processing purpose and other usage, I want to take screenshots of the 3D model at various angles. Following the Render-To-Texture capability, I noticed when my scene is not visible, the drawSceneContentWithVisitor: method only execute once rather than at every rendering cycle. For obvious reason, the CC3GLFramebuffer* won't get updated with new data, hence, I'm only able to take the initial screenshot.
Thanks.
In Cocos3D, you can render your 3D scene to an off-screen surface. See the CC3DemoMashUp addTelevision and drawSceneContentWithVisitor: methods for an example of how to do this.
What is important is that the 3D drawing environment has been established when you perform your drawing. The safest place to do this is inside your drawSceneContentWithVisitor: method. But if you want to render somewhere else, you need to invoke the CC3Scene open3DWithVisitor: and CC3Scene close3DWithVisitor: methods before and after rendering. See the implementations of the CC3Scene processInitializeScene and open methods for examples of how to do that.
To render your scene from multiple viewpoints, you need to add multiple cameras to your scene, and set the camera property of your drawing visitor appropriately to select a camera before rendering. See how this is done in the CC3DemoMashUpScene addTelevision and drawToTVScreen methods. The drawToTVScreen method also shows how to handle clearing the color and depth buffers of your surface.
Anyone have any idea how I can implement this? I'd like to have a function basically exactly like impoly in matlab or the "polygon sections" tool in imageJ, where you click to form a polygonal section and then each node can be adjusted, etc. I'd also like to have access to this function from Qt since I'm trying to make a gui for a small program I wrote.
Also, I'd like to avoid making calls to the matlab function because it's part of the image processing toolbox which isnt free. Thanks.
I think the best way to implement this is using the Qt Graphics View framework. Create a scene with an Item displaying your image in the back and add draggable Items on top representing the corners of your polygon.
Your selection tool should probably be a subclass of QGraphicsObject hosting the polygon corners as child items and a QGraphicsPolygonItem below the corners being updated whenever the user readjusts the selection. As QGraphicsObject inherits QObject, you can emit signals with a QPolygonF or QPolygon argument whenever the selection changes, informing other parts of your application
This demo should be a good example of the corner-adjust functionality you need.
Qt Pathstroke Demo
(uh well, the example implements the drawing and dragging of the control points from scratch.. I'm sure you can do it by using QGraphicsEllipseItem instead and react on their position changes)
I think you would need to code this yourself. There is an excellent example in the C++ GUI Programming with Qt 4 book (there's a PDF copy floating around online; I think it's legal) where they show you how to create a diagram with nodes and links. The chapter is called "Item-based rendering with Graphics View".
The basic idea is that you have some draggable nodes, which are QGraphicsItems with the ItemIsMovable flag set to true, and then some links that connect them, which are QGraphicsLineItems. All of these would go into a composite QGraphicsItem representing the ROI, and all of those would go into a QGraphicsScene, which would be displayed by a QGraphicsView.
Bottom line: there isn't a built-in copy of the MATLAB function, but all the tools are there for you.
Hy!
I'm building a top view 2D map, that it's objects are stored on the server.
The kind of objects are 10 and might be a photo, label, button, lists, mix of them or labels with tooltips.
The component must request the "areas" that are missing on screen.
An area is 1000x1000 px and is cached in flex.
To move in the map, will be like in google maps (drag-and-drop)
I should be able to have another list and move objects from one to another using drag-an-drop on objects. Ex.: I grab an objects from a list and I move it on this map, I release the mouse button and the item is placed there.
Now the problem is: I build a custom component for this trying to emulate the item renderer for performance and recyclage, implement drag-and-drop on objects and request the areas that are missing?
or
I extend the List component from spark and I add some features as multiple kind of itemrenderers and use recycle on them. Of course it must be able to request the missing areas on the screen and cache it's data.
Maybe create a custom layout is needed too.
What I need is something that must be really fluid, so the lighter this component is, the better!
Thanks for your help! (:
UPDATE:
*There will be not any object over another.
*I will not use hitTest on bitmaps because all bitmaps are wrapped in another component,as they,for now are itemrenderers.
Anyway I already begin to do this using a class that extends the SkinnableDataContainer and a custom layout. As the layout is not like a grid, is sparse, random items at diferent points(x, y).
Now I have this problem: https://stackoverflow.com/questions/4192934/how-to-get-the-localx-and-localy-relative-to-item-renderer-and-not-to-the-spark
Maybe you could use a combination of some of the following:
A canvas background (for layered components), looks fluid if using hitTest. Split up the canvas using constraintColumn/constraintRow for a grid layout.
Use hitTest (on items bitmapdata) for collision detection, and move the items to a new position (if items are dropped on top of one another).
Use a combo of mouseDown, mouseMove, mouseUP (example here) for drag n drop.
Use a 2D/3D indexed array to track the position of items (example here)