1. What are the future works in "Cardinality-constrained texture filtering" ?
Another possibility is that hardware designs will change to better support random access or even the access pattern of their method.. It may be possible to incorporate the current texel fetch behavior of GPUs in their optimization.. The idea of simultaneously optimizing basis functions and their coefficients for filter reproduction [ Gotsman 1994 ] has potential for producing even better results when combined with their idea of optimizing for which texels to use from different resolutions ; although the simultaneous optimization may be complex to solve.
read more
2. What are the contributions mentioned in the paper "Cardinality-constrained texture filtering" ?
The authors present a method to create high-quality sampling filters by combining a prescribed number of texels from several resolutions in a mipmap.. To find the best set of texels to represent a given sampling filter and what weights to assign those texels, the authors perform a cardinality-constrained least-squares optimization of the most likely candidate solutions and encode the results of the optimization in a small table that is easily stored on the GPU.. The authors present results that show they accurately reproduce filters using few texel reads and that both quality and speed scale smoothly with available bandwidth.. Their technique provides fine control over the number of texels the authors read per texture sample so that they can scale quality to match a memory bandwidth budget.
read more
