Project 4: Han Wang

README.md

@@ -3,11 +3,61 @@ CUDA Denoiser For CUDA Path Tracer
 
 **University of Pennsylvania, CIS 565: GPU Programming and Architecture, Project 4**
 
-* (TODO) YOUR NAME HERE
-* Tested on: (TODO) Windows 22, i7-2222 @ 2.22GHz 22GB, GTX 222 222MB (Moore 2222 Lab)
+Han Wang
+
+Tested on: Windows 11, 11th Gen Intel(R) Core(TM) i9-11900H @ 2.50GHz 22GB, GTX 3070 Laptop GPU
+
+|without denoizer|with denoize|
+|:-----:|:-----:|
+|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-19_01-52-06z.181samp.png" width="300" height="300">|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-20_23-54-52z.277samp.png" width="300" height="300">
+### Analysis
+Firstly, Among the tested output, since the denoize time will be affected by the various variables, I analyzed the denoize in the default condition given to us:
+800x800 resolution, 80 filter size. The output is: 4.645Ms
+
+
+Based on my observation, with the denoize feature implemented, the iteration amount needed for reaching the acceptable smooth condition was significantly reduced, leading to more efficient and time-saving results in the process.
+
+Also, based on my test I got the following graph among the resolution and runtime of denoize:
+
+|Graph1|
+|:-----:|
+|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/denoize_resolution.png" width="700" height="500">
+
+
+
+It is easy to see that while the resolution increased, the amount of that denoize needed was increased. It is also easy to understand: there are more pixels needed to estimate, and thus takes more time to denoize.
+
+Additionally, I get the following graph among the filter size and runtime of denoize:
+
+|Graph2|
+|:-----:|
+|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/resolution.png" width="700" height="500">
+
+It is easy to see from the graph, that the runtime needed increased a little bit while the filter size increased. But the increase is rather small. For the small wave in the graph, while the filter size increases, I guess it is because the increased amount is small and even hidden.
+
+
+I recorded the visual output for different filter sizes and get the following graph:
+(For comparing, the iterations was set to 100)
+
+
+|0|10|20|30|40|
+|:-----:|:-----:|:-----:|:-----:|:-----:|
+|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-28-26z.100samp.png" width="200" height="200">|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-33-12z.100samp.png" width="200" height="200">|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-33-40z.100samp.png" width="200" height="200">|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-34-03z.100samp.png" width="200" height="200" >| <img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-34-22z.100samp.png" width="200" height="200" >
+
+|50|60|70|80|90|
+|:-----:|:-----:|:-----:|:-----:|:-----:|
+|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-34-56z.100samp.png" width="200" height="200">|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-35-21z.97samp.png" width="200" height="200">|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-35-21z.100samp.png" width="200" height="200">|<img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-35-44z.100samp.png" width="200" height="200" >| <img src="https://github.com/Ibm510000/Project4-CUDA-Denoiser/blob/base-code/img/cornell.2023-10-21_00-36-01z.100samp.png" width="200" height="200" >
+
+
+It is easy to see, while the filter size increases, the output becomes gradually clear and easy to see.
+
+
+The next thing is material. I tested the diffuse and reflection, there are no significant runtime differences in denoise between them. In my opinion, the material might affect the basic part of ray tracing, the denoize is likely a post-process approach and won't be affected. 
+
+
+In different scenes, the denoizing effect will change based on the condition. For example, in the Cornell-celling scene, since there is more light there, the overall scene is brighter and can easily reach the effect Cornell scene needs higher iterations to get. Thus, it can provide better results with a small filter size. For more complex and less light scenes, we will need more iterations to get acceptable output. Otherwise, we need to increase the filter size.
+
+
 
-### (TODO: Your README)
 
-*DO NOT* leave the README to the last minute! It is a crucial part of the
-project, and we will not be able to grade you without a good README.
 

scenes/cornell.txt

@@ -52,7 +52,7 @@ EMITTANCE   0
 CAMERA
 RES         800 800
 FOVY        45
-ITERATIONS  5000
+ITERATIONS  100
 DEPTH       8
 FILE        cornell
 EYE         0.0 5 10.5

scenes/cornell1000.txt

@@ -0,0 +1,117 @@
+// Emissive material (light)
+MATERIAL 0
+RGB         1 1 1
+SPECEX      0
+SPECRGB     0 0 0
+REFL        0
+REFR        0
+REFRIOR     0
+EMITTANCE   5
+
+// Diffuse white
+MATERIAL 1
+RGB         .98 .98 .98
+SPECEX      0
+SPECRGB     0 0 0
+REFL        0
+REFR        0
+REFRIOR     0
+EMITTANCE   0
+
+// Diffuse red
+MATERIAL 2
+RGB         .85 .35 .35
+SPECEX      0
+SPECRGB     0 0 0
+REFL        0
+REFR        0
+REFRIOR     0
+EMITTANCE   0
+
+// Diffuse green
+MATERIAL 3
+RGB         .35 .85 .35
+SPECEX      0
+SPECRGB     0 0 0
+REFL        0
+REFR        0
+REFRIOR     0
+EMITTANCE   0
+
+// Specular white
+MATERIAL 4
+RGB         .98 .98 .98
+SPECEX      0
+SPECRGB     .98 .98 .98
+REFL        1
+REFR        0
+REFRIOR     0
+EMITTANCE   0
+
+// Camera
+CAMERA
+RES         1000 1000
+FOVY        45
+ITERATIONS  5000
+DEPTH       8
+FILE        cornell
+EYE         0.0 5 10.5
+LOOKAT      0 5 0
+UP          0 1 0
+
+
+// Ceiling light
+OBJECT 0
+cube
+material 0
+TRANS       0 10 0
+ROTAT       0 0 0
+SCALE       3 .3 3
+
+// Floor
+OBJECT 1
+cube
+material 1
+TRANS       0 0 0
+ROTAT       0 0 0
+SCALE       10 .01 10
+
+// Ceiling
+OBJECT 2
+cube
+material 1
+TRANS       0 10 0
+ROTAT       0 0 90
+SCALE       .01 10 10
+
+// Back wall
+OBJECT 3
+cube
+material 1
+TRANS       0 5 -5
+ROTAT       0 90 0
+SCALE       .01 10 10
+
+// Left wall
+OBJECT 4
+cube
+material 2
+TRANS       -5 5 0
+ROTAT       0 0 0
+SCALE       .01 10 10
+
+// Right wall
+OBJECT 5
+cube
+material 3
+TRANS       5 5 0
+ROTAT       0 0 0
+SCALE       .01 10 10
+
+// Sphere
+OBJECT 6
+sphere
+material 4
+TRANS       -1 4 -1
+ROTAT       0 0 0
+SCALE       3 3 3

scenes/cornell1100.txt

@@ -0,0 +1,117 @@
+// Emissive material (light)
+MATERIAL 0
+RGB         1 1 1
+SPECEX      0
+SPECRGB     0 0 0
+REFL        0
+REFR        0
+REFRIOR     0
+EMITTANCE   5
+
+// Diffuse white
+MATERIAL 1
+RGB         .98 .98 .98
+SPECEX      0
+SPECRGB     0 0 0
+REFL        0
+REFR        0
+REFRIOR     0
+EMITTANCE   0
+
+// Diffuse red
+MATERIAL 2
+RGB         .85 .35 .35
+SPECEX      0
+SPECRGB     0 0 0
+REFL        0
+REFR        0
+REFRIOR     0
+EMITTANCE   0
+
+// Diffuse green
+MATERIAL 3
+RGB         .35 .85 .35
+SPECEX      0
+SPECRGB     0 0 0
+REFL        0
+REFR        0
+REFRIOR     0
+EMITTANCE   0
+
+// Specular white
+MATERIAL 4
+RGB         .98 .98 .98
+SPECEX      0
+SPECRGB     .98 .98 .98
+REFL        1
+REFR        0
+REFRIOR     0
+EMITTANCE   0
+
+// Camera
+CAMERA
+RES         1100 1100
+FOVY        45
+ITERATIONS  5000
+DEPTH       8
+FILE        cornell
+EYE         0.0 5 10.5
+LOOKAT      0 5 0
+UP          0 1 0
+
+
+// Ceiling light
+OBJECT 0
+cube
+material 0
+TRANS       0 10 0
+ROTAT       0 0 0
+SCALE       3 .3 3
+
+// Floor
+OBJECT 1
+cube
+material 1
+TRANS       0 0 0
+ROTAT       0 0 0
+SCALE       10 .01 10
+
+// Ceiling
+OBJECT 2
+cube
+material 1
+TRANS       0 10 0
+ROTAT       0 0 90
+SCALE       .01 10 10
+
+// Back wall
+OBJECT 3
+cube
+material 1
+TRANS       0 5 -5
+ROTAT       0 90 0
+SCALE       .01 10 10
+
+// Left wall
+OBJECT 4
+cube
+material 2
+TRANS       -5 5 0
+ROTAT       0 0 0
+SCALE       .01 10 10
+
+// Right wall
+OBJECT 5
+cube
+material 3
+TRANS       5 5 0
+ROTAT       0 0 0
+SCALE       .01 10 10
+
+// Sphere
+OBJECT 6
+sphere
+material 4
+TRANS       -1 4 -1
+ROTAT       0 0 0
+SCALE       3 3 3