Close and far Future Work
- Modify the behavior of the system (before modify the interface):
- Constraint detector
- Automatic time stop
- Update the kernel:
- Include torsion in the fairing curve
- Correct the computation of w terms
- Speed up the wavelet transformation
- Check the height field when increasing the resolution
- Finalize the new interface:
- Zoom in the left window
- Add scrolling in the both side
- Complete the status bar informations
- Adjust the sensibility of the mouse
- implement the new ©P.Bill interface
- Future work – from closest to furthest:
- Relighting the scene : ideas
- Find an illumination model that works for one particular image (Jelly Bean)
- Set the conditions to detect automatically where the light comes from
- Specular properties fixed ?
- Geometric properties fixed ?
- use shadows, reflects, self-shadowing ?
- Use noise and other post-treatements to make the results good-looking
- Maybe use a bump-map system to get more realistic results
- Automatic subdivision
- Silhouette curve
- Handling large images
- Handling perspective
- Filling the holes:
- Solution 1: extending the current system with 3 different functions instead of one.
- Solution 2: implicit functions, metaballs...
August work : illumination
- Given the position of the light source and the geometry of the system:
- solve for kd as a function without assumptions
- solve for kd as a piece-wise constant function (hard constraint)
- solve for kd as a piece-wise constant function (smooth constraint)
- Given only the geometry of the system,
- solve for kd as a piece-wise constant function (hard constraint)
- solve for kd as a piece-wise constant function (smooth constraint)
- Find a way to add specular properties
- Modify the segmentation of kd domain of defintion to make it more automatic
- Automate the computation of the geometry from the illumination and the specular properties
- Add some residual models and after-effect filters to render the scene in a good-looking way
Next Week
- make an interface to change the light direction
- make an interface to view the result of kd value
- compute the kd as a simple function (first part)
- compute the kd as a simple function (first part)
- See if it is possible to code the new illumination with OpenGl, otherwise
do it by myself