Experimental research topic and set up a Timeline (IMPORTANT node in my research)

My new topic is Unity Graphic Rendering based on Shader and AI System

 Before I came up with this idea, I've been looking for the specific topic for a while. The way should be a formal research method, starting from raising question. So the first question I raised for myself is


1.What can shader do as innovation?

Not long ago, I was considering to make a better graphic display, but later, nah, a lot people had done it already. What really inspired me was a lecture talks about AI 3D generator (It's in Chinese so I don't put the link there). Basically, if the group wants to build a forest, it's not possible for them to build every tree, which costs too much time. So they give AI a reference model, and AI can generate new trees with slightly different angels or length of branches and leaves.


From this, I was thinking if the same process could be applied in shader. To prove my guess, I need to do some research in this field, so I raised the second question


2.How does AI normally work?

Just like the model generator, AI ususally needs a reference to do the simulation work. So what I am going to do is to have a deeper understanding about shader structure so that I can write the corresponding algorithm to be applied with AI.

3.Is there any examples for combining AI and graphics?

NVIDA published 20 series graphics card two years ago. One of the new function is called DLSS, deep learning super sampling. It uses AI to automatically fix the errors come from other part of rendering which shows in the wrong position. That's the example I found, but I am not really sure it could be a example for my research as well. It still remains to be digged.



3.Significance?

It reduces people's work as most of AI do, and it also allows the use of shader more flexible. Besides, since AI is engaged, it's very likely to have something unexpected, I have no idea what it will be like, but worth to try.


Summary:
It's still a very general research topic, I don't know what kind of visualization I should be working on, that's mainly because I know too less about both shader and AI system. Hopefully, it could be a shader generator which changes the graphic effect or materials according to some certain conditions(environment, for example), as to how it could be practically applied, no clues yet.


Anyway, I set up the schedule for the following weeks.
Week 6(this week):

1. Keep shader learning, at least one practice work per week.
2. Go find some simplest AI package regardless of function, try to use it myself.
3. Research on more examples on the AI and graphics combination.

Week7:

1.Keep shader learning, at least one practice work per week.
2.Analyze the codes in AI package, try to change the 'reference' part.
3.get more information about shader structure, what can be extracted as the reference?

I'm not gonna plan all weeks cuz that's kinda unrealistic, I will definitely take more time than I though on these works above and I might have some changes about the topic. I will have another schedule later.














Comments

Post a Comment

Popular posts from this blog

Thesis Progress - Noise Map

Image
 In another around of research, I found a toolkit and try to use it for my project.  Resource:  https://github.com/mtwoodard/TextureGenerator It's a 3D texture generator designed for raymarching clouds. As we know, the process of my project includes the steps of asset shape then blurred by noise map. Both of them are based on 3D texture. The initial idea was to apply perlin and worely noise, cuz that's the method introduced by many paper. But now, I got this tool which makes it easier.  From last blog, I already had a framework that can drive the 3D texture for some extent. And the noise map makes it pretty close to the final result. Here's what it looks like and it makes me HAPPY. There's an example in the toolkit , I applied it with my 3D texture script. It's not that clear now, then I juggled with some parameters on Blur Material.  Transparency: 0.4 OpacityCorrection: 2.4 Then the outcome shows some really nice work. And each of those plane is attached to tex...
Read more

Thesis Progress - Schedule Check

Image
According to the timetable, the goal of this quarter is to finish the first two parts which are the function experiment for thesis project. So far,  the first method I came up with is going through 40% around, still need some more learning on coding to realize modification. At the same time, I'm still trying to find the other ways that might be better or easier, which could replace the current method. The way I make it should be covered by my capability as much as possible.  Other than that, I'm also considering whether I should have a comparsion between my method and traditional way of making 3D texture. It's mostly from the thesis perspective that I need to show the advantage or benefit of my own method. That requires more specific arrangement on timeline. 
Read more

Thesis Progress - Procedural 3D texture

Image
 According to tutorials, there's some outcome. The function realized plane creating, which would be saved later for my further using.  The problem so far is that the sphere generating function is way too complicate to modify myself, and it closely related to the planes, so it might cost more time on learning.  There's another easier example, what I can understand is the switch from object to worldspace. Once the cube is out of range, the texture could not sample on the object accurately. But not sure if I would use this later, but good to know what's going on in codes.  Resources archive:  https://www.youtube.com/watch?v=FUpsRzricPA&t=3s            The second example is from unity documentation that introduces the usage of 3D texture, this part could be as the framework of raymarching that allows to modify the shape possibily. This part will be put behind currently.  Resources archive:  https://docs.unity3d.com/Manual/...
Read more