Submission Format. Electronic submission on Canvas is mandatory. Submi t in a zip file, a single pdf file containing:
• the source code,
• brief of explanations of what was done, • images and screenshots of the model.
Also include the final ply model.
The goal is to obtain a voxel-based reconstruction of the scene using the provided silhouettes as inputs.
1
Part 2. The goal of this part is to identify surface points that should be included in the output 3D model. You can represent faces of voxels by one point each. The output should be in ASCII ply format. Use the included ply file as a sample for how to create such as file. The number after element vertex is the total number of vertices and has to be correct for the model to be displayed correctly. Each vertex is represented by three floating point numbers for the coordinates and three unsigned characters for the RGB colors. In this part, it is acceptable to have all points be of the same color. Models can be visualized using Meshlab, which is available at http://meshlab.sourceforge.net/ and works on Windows, MacOSX and Linux.
Part 3. Try to color the model by selecting RGB values from the images. Using one image per point is perfectly acceptable, but the criteria for choosing it have to be described in the report. It will be helpful to determine a vector for each vertex that points outward from the occupied volume to the cameras.
Acknowledgement. The input images, silhouettes and calibration parameters have been generated and made available by the Morpheo team at INRIA, Grenoble, France.
2
