Fixing Bugs

The last 2 weeks have been spent fixing bugs and working on making sure that each piece of the algorithm is up to date. Some of the bugs that I’ve worked on have included:

• An issue where uv coordinates were being compared instead of position coordinates, which would create problems with accessing correct texture points. The solution to this was to add an additional set of position data to map out each source position so that position data could be easily compared in 3D instead of in 2D.
• An issue where 3D textures were being muddled, causing strange behaviour when fetching texture coordinates. For now, a set of 3 2D textures is being used, but this will soon be swapped to an array of textures.
• An issue where position data was being used in 2D instead of in 3D.

Other issues that have been noted, but not addressed yet, include:

• An issue where patches are not being selected, but single colours are. I believe that this bug will require adding additional UV points are mapped to a texture in order to select different pieces of the source texture.

Over the last few weeks, I have also worked on updating some of the models, textures, and scripts used. I am now working with the Stanford bunny model, and I have also cleaned up and imported the humanoid model used by the StyleBlit developers. Each model needed some extra tweaking, whether due to strange size and orientation, or due to lack of uv coordinate data. This was added in Maya.

Other work has included updating my textures, fetching new resources, and working on the implementation of the script to approximate the data more closely.