CS-GY 6533 A Interactive Computer Graphics
Assignment 3: 3D Scene Editor
Handout date: 10/28/2019
Submission deadline: 11/18/2018, 23:59 EST
Demo date: 11/22/2019
This homework accounts for 17.5% of your final grade.
Goal of this exercise
In this exercise you will implement a 3D editor that allows to prepare 3D scenes composed of multiple
3D objects.
Eigen or GLM
In all exercises, you will need to do operations with vectors and matrices. To simplify the code, you will
use Eigen or GLM. Have a look at the ”Getting Started” page of Eigen As well as the Quick Reference
page, and ”Manual” document of GLM to acquaintain yourselves with the basic matrix operations
supported.
OpenGL
In all exercises, you will use OpenGL 3.3 with GLSL version 150 or newer versions.
Submission
You must submit a zip file with all your code and libraries (glew, glfw, Eigen or GLM) used in this
assignment. The zip file must be submitted using the NYU Classes system as you did for Assignment
1 and 2.
Try to maintain the same directory organization as the starter code, so you don’t need to change the
CMakeLists.txt file.
Don’t forget about the README file/report.
1 Mandatory Tasks
For each task below, add at least one image in the readme demonstrating the results. The code that
you used for all tasks should be provided.
Claudio Silva and Jonathas Costa - Based on Daniele Panozzo’s Original Notes 1
CS-GY 6533 A Interactive Computer Graphics
1.1 Scene Editor
Implement an interactive application that allows to add, edit, and delete 3D meshes. The scene should
always contain at least one light source. New objects can be added to the scene in three ways:
• The key ’1’ will add a unit cube in the origin
• The key ’2’ will import a new copy of the mesh bumpy cube.off, scale it to fit into a unit cube
and center it on the origin
• The key ’3’ will import a new copy the mesh ’bunny.off’, scale it to fit into a unit cube and center
it on the origin
Note that you can have multiple copies of the same object in the scene, and each copy can have its
own position, scale, and rotation. For this exercise, all transformations MUST be done in the shader.
The VBO containing the vertex positions of each object should be uploaded only once to the GPU.
1.2 Object Control
Clicking on a object will select the object, changing its color. When an object is selected, it should be
possible to translate it, rotate it around its barycenter, and rescale it without changing its barycenter.
All these actions should be associated to keyboard keys (and the choice of keys should be detailed in the
readme). Each object also has a rendering setting associated with it, which can be one of the following
three options:
1. Wireframe: only the edges of the triangles are drawn
2. Flat Shading: each triangle is rendered using a unique color (i.e. the normal of all the fragments
that compose a triangle is simply the normal of the plane that contains it). On top of the flat
shaded triangle, you should draw the wireframe.
3. Phong Shading: the normals are specified on the vertices of the mesh and interpolated in the
interior. The lighting equation should be evaluated for each fragment.
Claudio Silva and Jonathas Costa - Based on Daniele Panozzo’s Original Notes 2
CS-GY 6533 A Interactive Computer Graphics
To compute the per-vertex normals you should first compute the per-face normals, and then average
them on the neighboring vertices. In other words, the normal of the vertex of a mesh should be the
average of the normals of the faces touching it. Remember to normalize the normals after averaging.
When an object is selected, it must be possible to switch between the different rendering modes by
pressing three keys on the keyboard.
1.3 Camera Control
Add the possibility to translate the position of the camera (similarly to the previous assignment), but
in this exercise the camera should always point to the origin. It should be possible to move it around,
but the camera should always face the origin.
Implement both a orthographic camera (similar to the one that you used for Assignment 2, but in
3D) and a perspective camera. The cameras should take into account the size of the window, properly
adapting the aspect ratio to not distort the image whenever the window is resized. All functionalities
should work after resizing the window, including object selection and editing of the scene.
Optional Tasks
These tasks are optional. The first task is worth 2% of the final grade, the second 3%, and the third
task 2%. The optional points are added to the points of the other exercises, but the total sum of points
that you gain with exercises cannot be more than 80%.
1.4 Animation Mode
Add the possibility to draw a Bezier curve for each object in the scene. This should be done by having
at least one key that allows to add a new control point and three keys to translate the point around.
It is not necessary to have a UI that allows to edit the curve. Whenever the button space is pressed,
Claudio Silva and Jonathas Costa - Based on Daniele Panozzo’s Original Notes 3
CS-GY 6533 A Interactive Computer Graphics
each object should start to move following its associated curve, i.e. it’s barycenter should follow the
curve. The animation should last 5 seconds, and at the end all objects should go back to their previous
state. The curve should be evaluated using De Casteljau’s algorithm (See the course’s textbook, pages
390-392).
1.5 Export in SVG format
Add the possibility to export the scene currently drawn on the screen in SVG format https://en.
wikipedia.org/wiki/Scalable_Vector_Graphics. The exported SVG should be compatible with
https://inkscape.org/. The triangles should be rendered only using flat shading and they should
be deformed according to the current view (orthographic and perspective should both be supported).
Hidden triangles should not be rendered (to check for hidden triangles, you can cast three rays to its
vertices, and if any vertex is not visible you will simply skip the triangle).
1.6 Trackball
Use a trackball to control the camera. This can be achieved restricting the movement of the camera on
a sphere centered on the origin. The easiest way to do it is to parametrize the sphere using spherical
coordinates, and to assign keyboard keys to move the camera on the sphere. The camera should always
look at the origin.
Claudio Silva and Jonathas Costa - Based on Daniele Panozzo’s Original Notes 4