One thing... as per table, chairs seating heights seems very low. I sitting position only heads will be above the table height.
@rombout I can see how doing animations can get it into your workflow to a point you use it for stills as well... video is just a lot of stills in the end. They do tend to pack more power - one example is the new DEEP Image editing which is super cool... see a glimpse of it in the video
Shown in this forum thread : http://www.ronenbekerman.com/forums/finished-work/6210-lakes-house.html
@camillelymer @rombout Thanks for explanation and of course the banana! I think this is a bit over my head. I havent really done animations or moving images. Perhaps ill get their eventually... PS i see a couple objects from that site, i got the link to that site last week. some nice objects for download
Here is a screen shot of an extreme close-up of my scene, showing the 3 deep pixels along an edge.
It has something to deal with aliasing. Let's take a pixel between your foreground and background. Let's say you want to put a 3d banana between those in nuke (a strawberry works also :) ) .
Nuke can't really know how to deal with this pixel as it belongs to both foreground and bakground (aliased). So you have to give him a mask (multimatte for instance) to reveal cleany your banana. Now even harder: Imagine you want a banana 50% transparent. Nuke can't output a clean pixel because the pixel can't store a foreground sample, a banana transparent sample, and a background one. It will roughly calculate an average, wich can be dirty especailly with animations.
BUT, with deep image, render engine can store many pixels in one (called "deep" pixels).
To see that you have a node in nuke called "deep sample", wich will show the number of deep pixels, with color, opacity, position for each 2d pixel.
Haven't found explicit video on the web showing this nuke's node in action, so you will have to try.
You can find a banana model here:
@camillelymer @rombout @ronenbekerman I did a quick peek in how it works and what it stands for. Didnt really get the part about those artifacts. If all the maps are clean why would one have artifacts???
Deep compositing seems very powerful, just checked some examples. SOrt of a layer system for video but also with 3 dimensional aspects
Proud to see you talk about my work here! :)
In this case after effects doesn't add really something more than photoshop (Moreover since "looks" plugin exists for photoshop as well).
But as ronen said, when softs for movies is part of your workflow, you'd rather keep working with those even if you are doing still frames.
Deep compositing is indeed very usefull when you have to deal with many differents footages (Plate, 3d, Vfx, object replacement), but it can also helps you reduce edges artifacts. That is valuable for movies and stills alike.
Lasse Rode takes Unreal Engine for a good spin as he strives for photorealism using a real-time tool. He does this... more