Lab 1

Studio Lighting

Due: Thursday, October 26, 2016 at 23:55

Start Early!!!

This lab was modeled after the assignment 1 of Stanford's Image Synthesis class.
Individual Effort:
No team participation is really encouraged in the case of the homework or the labs.
Academic Misconduct
Late Submission Policy

Objectives

This assignment helps you familiarize yourself with PBRT, the physically based rendering toolkit, and prepare for the future lab assignments. You will experiment with different scene settings and parameter combinations.

Step 1: Building PBRT

Please download the pbrt-v3 code from the PBRT website in the Downloads section. A large collection of pbrt-v3 scenes are also available in the Scenes section. If you download the repository snapshot as the .tar.gz file, you might need to receive the scene files from the repository with the following commands: 

git checkout . && git pull
You can also download individual models from Benedikt Bitterli's rendering resources web page.

To create the makefile or Visual Studio project, you need to install CMake first. You can find detailed instructions for setting up and compiling the PBRT here. Alternatively, you can follow these instructions:

To be able to call the pbrt command from any folder, you can add it's address to the path environment variable. On Linux/MacOS, it is done by: 

export PATH=/path/to/pbrt/bin/:$PATH
You can add this command to your .bash_profile on Linux/MacOS. On Windows, you need to go to the System's menu in Control Panel and add the folder to your PATH environment variable. For the rest of this lab instructions, I assume that you have done this step. If you have skipped this step, you will need to specify the full address for calling the pbrt command.

Step 2: Rendering Your First Image

Once the compilation is done, you can render your first image. The starter files can be downloaded here. After you unzip the folder, you can render the image with this command: 

pbrt lighting.pbrt
After the rendering is done, it will print out the rendering statistics and the image will be stored in lighting.exr file. The name of the output image is specified in the input lighting.pbrt file, as a parameter in the Film definition. You can pass different filenames with other extensions (e.g. png or jpg) and pbrt will automatically determine the encoding from the file extension. The EXR image has a high dynamic range (HDR) format and stores linear radiance values, which can have a very large range, beyond the range that your screen can handle. Several image processing/viewing tools can do the conversion automatically (e.g. OS X's Preview, Adobe Photoshop, GIMP 2.9.2+, OpenEXR exrdisplay). PBRT has a built in tool for this conversion, called imgtool. You can convert the image to the png format using this tool: 
imgtool convert lightings.exr lighting.png

Please always convert your output images to png using imgtool before submitting it, unless specified otherwise.

A correctly rendered image is shown above on the left. The scene setting is shown on the right side, where the camera is pointed directly at the model of a face. A spotlight shines on the face from behind the camera and slightly off right, casting hard shadow on the wall behind the model. The location of the camera and spotlight are illustrated in the image.

Before moving on to the rest of this assignment, please open the lighting.pbrt file with your favorite text editor and go through the model and the comments around them. Try to understand the camera parameters and the head model. Keep in mind that pbrt uses a left-handed coordinate system. It is also highly recommended that you take some time to go over Camera, Lights, and Area Lights in the pbrt-v3 file format manual before you start modifying the scene file.

Changing Rendering Parameters

The image properties are defined in the scene file under the Film definition: 

Film "image" "string filename" ["lighting.exr"]
    "integer xresolution" [300] "integer yresolution" [300]
Beside the filename and extension, you can also change the image resolution with integer xresolution and integer yresolution parameters. Notice that the parameter type is always specified before the name in pbrt scene file format.

The image resolution has a direct effect on the rendering speed. So, you can reduce the resolution in your tests if you want faster rendering when playing with parameters but you should submit your images with the original resolution of 300x300

Another factor that affects the rendering speed is the number of samples (eye rays) that pbrt uses to compute the value of each output pixel. The Sampler definition in the scnee file is: 

Sampler "halton" "integer pixelsamples" [4]
Please submit your images with at least 4 samples per pixel unless mentioned otherwise. In your own experimentations, you can change this parameter to increase the rendering speed.

Step 3: Lighting the Scene

For the rest of this assignment, we will try several different lighting configurations.

Configuration 1: Spotlight from Below

In this configuration, we will move the spotlight so that it points up the face from below. To do so, you will need to modify the spotlight parameters in lighting.pbrt file, in the spotlight section: 
LightSource "spot" "color I" [50 50 50] "point from" [-0.5 0 4.7] 
    "point to" [0 0 0] "float coneangle" [60]
Your task is to find the right positions for point from and point to parameters of the spotlight to create the lighting above.

Configuration 2: Area Lighting from the Right

The light generated by a spotlight originates from a single point. Light that is originated from a single point or a small area creates hard shadows that photographers generally find visually objectionable. Area lights can emit light from a larger region in space, resulting in a softer lighting of the object and a shadow with penumbra. The pbrt scene file contains the following definition of an area light, which is originally commented out: 
AttributeBegin  
  AreaLightSource "area" "color L" [10 10 10]
  # use camera coordinate system [optional]  
  CoordSysTransform "camera"  
  # adjust light source position  
  Translate 0 0 -2  
  Rotate 90 1 0 0  
  # define the shape of the arealight to be a disk with radius 1.5  
  Shape "disk" "float radius" [1.5]  
AttributeEnd

Light from this source is emitted from a disk of radius 1.5. Comment out the light from configuration 1 and modify the area light above to generate the picture below:

Configuration 3: Two Light Sources

Photographers usually use a reflecting surface on the opposite side of the light to prevent the object's shadow on itself and have a more uniform lighting of the model. The second source of light on the model is referred to as fill lighting. We model this effect in this assignment by using two light sources. Illuminate the model with a spotlight from the position indicated in the illustration below. Then use a large area light on the right of the model to give the effect of fill lighting on the dark side of the face. Render the scene with and without the fill lighting to observe the visual differences.

Configuration 4: Many Light Sources

In practice, photographers use many lights to illuminate a subject so that the resulting image conforms precisely to their preferences. Try to adjust the lighting condition to recreate the scene below, in which four light sources are used to create a visually pleasing effect. You will also need to move the camera for this configuration. The LookAt directive specifies the position of the camera in world space, the location it is pointed at, and the "up" direction of the camera. Camera settings are defined in the scene file by the following two lines: 

LookAt 0 0 4.5  0 0 0   0 1 0
Camera "perspective" "float fov" [22]

To help you find the right configuration of the 4 lights, the following 4 images show the separate effect of the four light sources:
The main light on the side of the face, away from the camera.
The fill light close to the camera on the opposite side from the main light.
The accent light shines towards the upper part from the back of the subject. Normally, it is used to highlight the texture of hair; or lack of hair in our case.
The background light separates the subject from the background. It is placed behind the subject and to one side, facing toward the backdrop.

Configuration 5: Realistic Lighting

Manually placing lights is a time consuming process and not very efficient in recreating the real environment. A popular technique to overcome this issue is to use a HDR "environment map" as a light source for the rendering. The environment map has captured light from all directions in a real-world environment and can simulate a real world lighting for our model. The following environment map is provided for you in the assignment's zip file, at textures/doge2_latlong.exr:

The necessary configuration line is already in the scene file as: 

LightSource "infinite" "string mapname" ["textures/doge2_latlong.exr"]
Uncomment out the environment map light and comment out all other light sources to generate the following image:

You will need to increase the pixelsamples of the Sampler significantly to get an image with a small amount of noise. Please use at least 64 samples per pixel for this configuration. This brings up an interesting trade-off that you can see everywhere: We were able to drastically simplify the amount of work that needs to be done by humans by making the computer work harder.

Configuration 6: Realistic Model and Materials

We have a more realistic model of the head geometry at geometry/head.ply, which can be textured as a human face by an image from textures/head_albedomap.png in the assignment's zip file. To change the model with the realistic model, comment out the "Head model" section in the scene file and uncomment the "More realistic head model" section. You might need to increase the pixelsamples again. You can also change the maxdepth parameter of the Integrator to capture the reflections of the light on the model itself. The following image was created using 4096 samples per pixel with maximum depth of 5. Since it might take a long time to render with these parameters, you can submit an image with any settings that has more than 64 samples per pixel and maximum depth greater than 3.

Step 6: Submission

Be sure to double check your final submission by unzipping it in another directory on a computer in the PC LAB and testing it. (Especially for last minute submissions.) A project that doesn't run will lose more points than one that is one day late. Be sure to submit any comments or remarks in a README.TXT file. Submit your files on Moodle here. You only need to submit your solutions for the configurations in Step 3. Submit everything in a ZIP file that contains: Note that all files should be contained in a single ZIP file.

We expect a good student to have to work approximately 8 hours on this assignment.

Academic Honesty