It;s been a couple of weeks since the HDR Image post, as I've been spending a lot of time working out how excactly Maya works along with getting to grips with how all the features I will need for the final project work, so here's a full break down of the Maya knowledge I will need to ensure I can achieve the desired result in the final comp.
With an HDR Image created and the 3D Camera Projection setup in Nuke it’s time to bring the scene into Maya.
Match Moving
To do this it’s a simple process of bringing the .exr file from Nuke into the Scene’s folder that is set up in Maya through the project settings.
This will result in the camera data, along with the grid planes that have been created in Nuke to be brought into Maya and will look a little be like the image below, there will also likely be a need to reduce the size of the points from the point clouds to ensure they are not as big as the originally appeared.
Once this has been done, and you have grouped and named the different planes, camera and point clouds you should also ensure that the project size is set to the same as that of the original footage that you are using for the scene.
Once the project size is set to the same size as the footage, the need to create an image plane using that footage is needed to bring in the footage to Maya. This is done by creating and image plane from the Nuke camera in the camera shape settings.
Skydomes
Following on from this, you will need to import the HDR Image into the skybox into Maya, this will ensure that lighting is present in the scene. From here you should ensure that you go into colour management to enable the colour spacing of the HDR Image to Raw, ensure that all lighting and future rendered reflections work within the scene. It’s also important that the camera setting for the skybox is set to 0 to ensure that the image plane is not distorted by the skybox image – this will not impact on the lighting.
Next week we’ll be looking at how to edit and add Arnold Renders and AOV’s to our 3D assets before taking the Maya Rendered images back into Nuke to complete a demo comp.
Arnold Renders
When applying Arnold Render meshes and surfaces to 3D assets, image based lighting is important to ensure that shadows are rendered out. At this point, the camera visibility should be reduced to zero to remove the image while keeping the lighting that is generate by the HDR Image. Scaling and setting the assets into the scene is also required when setting up CG elements against live action footage to maintain a level of realism. Planes can then be brought in, if not generated in Nuke, to allow shadows to be added.
Following this set up, adding Arnold AI surfaces enable the addition of these shadows, reflections, and other such effects to be brought into the CG assets. Many of these are created using the lighting that is generated from the HDR Image used for the skydome, resulting in a natural lighting based on the lighting from these images.
To reduce ‘noise’ from the assets, the render settings -> Arnold render can be used to tweak the sampling layers to clean up the image by reducing the ‘noise’ from the assets. When doing this, the settings for the Alpha channels may become altered and bring in the Alpha’s from the background image. To rectify this before rendering, the image plane display mode should be set to none, this will ensure that the camera image plane’s Alpha channel is not brought in to the render shot, ensuring that the Alpha channels that are being used for the shadows on the assets.
In terms of the CG Asset’s actual texturing when adding Arnold layers there are 2 options. In this weeks lecture we were shown how to add these using the attribute editor, however I did additional research in to the use of the Hypershader tool which provides further control over how you can plug individual layers into an object to create a specific look at feel for the asset in Maya.
In terms of the attribute editor, the process is relatively simple, select the asset and initially create an aiStandardsurface. This allows Maya to render the asset with a range of different sample layers by using specific files which is useful for assets which have specific sample layers that are ordered and organised with specific sample layers. When bringing in each of these images using the file select tool in the attribute editor, it is important to set the colour spacing to RAW to ensure that the file provides as much data as possible for the render layers.
Once these steps have all been taken, and adjustments to each layer have been made in terms of exposure etc, the sequence can then be rendered out from Maya using the Arnold renderer. To maximise the ability to edit each of these levels in Nuke however, it is important to set the render layer settings for each individual asset and to also separate out the CG assets from the shadows using absolute overrides in terms of making the CG asset not visible, and to also remove unnecessary channels that are on the CG asset and not the shadow. The main reason for this is to reduce the render time of the shadow layer by giving Maya less data to need to render.
Further Reading & Research
Moving from Nuke to Maya has been a steep learning curve, as I’ve never used Maya before. To that end I spent some time this week looking at the basics of Maya as a software package to be better equipped with what I need to effectively and appropriate use the software ahead of my final composition project.
Using this YouTube tutorial, I learned how to navigate the software effectively, and how to use some of the more basic features. This supplemented the information we were taught in this week’s lecture. https://www.youtube.com/watch?v=kYQ98Q5UmNo
On top of learning the basics of Maya to ensure that I am able to effectively use the software, I did some further reading around the use of the Hypershader tool as a way to create and apply different Arnold AOV’s and sample layers using the following YouTube tutorials. https://www.youtube.com/watch?v=bx0PVWThTYY & https://www.youtube.com/watch?v=peuVIN1H9Rs
After reviewing these tutorials, and having looked at the large variety of sample layers for my assets, I intend to use the hypershader tool to create the AOV’s for at least one, if not both of my CG assets in Maya which will hopefully allow me to organise and create and appropriate set of Arnold shaders and AOV’s ahead of the final Nuke comp.
Comments