Shading is an important feature for the comprehension of volume datasets, but is difficult to implement accurately. Current techniques based on pre-integrated direct volume rendering approximate the volume rendering integral by ignoring non-linear gradient variations between front and back samples, which might result in cumulated shading errors when gradient variations are important and / or when the illumination function features high frequencies. In this paper, we explore a simple approach for pre-integrated volume rendering with non-linear gradient interpolation between front and back samples. We consider that the gradient smoothly varies along a quadratic curve instead of a segment in-between consecutive samples. This not only allows us to compute more accurate shaded pre-integrated look-up tables, but also allows us to more efficiently process shading amplifying effects, based on gradient filtering. An interesting property is that the pre-integration tables we use remain two-dimensional as for usual pre-integrated classification. We conduct experiments using a full hardware approach with the Blinn-Phong illumination model as well as with a non-photorealistic illumination model.
Volume rendering is one of the key technique to display data from diverse application fields like medicine, industrial quality control, and numerical simulations in an appropriate way. The current main limitations are still the inadequate rendering speed and the limited flexibility of the most efficient algorithms. In this dissertation, we developed three new algorithms for the acceleration of direct volume rendering and volume deformation. The first algorithm consists on a first step, on the reimplementation of the existing preintegration volume rendering approach, where the gray values between two sampling points change linearly, by considering the correct not simplified volume rendering integral, i.e, considering the attenuation factor as well as the shading function during the precompuation process. On a second step, we extended our algorithm to quadratic and higher order polynomial model. The preintegration speed for linear model is increased by a factor of 10. The second algorithm accelerates shear warp and ray casting process. While acceleration techniques like space leaping and early ray termination are efficient when rendering volumes with most of the voxels are mapped either opaque or transparent, encoding coherence appeared more efficient for rendering semi-transparent volumes. It's an approach for coding empty regions to a coherency encoding that can describe regions where the opacity changes linearly. We reimplemented this technique using a volume graphics library (VGL). We improved it by using the preintegration technique to evaluate opacity and shading inside the coherent region. We achieved a speedup of up to a factor of 3. The third algorithm is for volume deformation. The applied technique is the ray deformation where the volume deforming and the volume rendering are incorporated into a single process. This is implemented in our approach, by combining the Free Form Deformation (FFD) and inverse ray deformation. Unlike the previous implementation, our opacity and shading calculation are based on the preintegration technique which allows us to handle different lengths of the sampled intervals in the polyline segments which approximate the deformed ray.