PROJECT TITLE :
$k^+$ -buffer: An Efficient, Memory-Friendly and Dynamic $k$ -buffer Framework
Depth-sorted fragment determination is prime for a bunch of image-based techniques which simulates complex rendering effects. It's conjointly a difficult task in terms of your time and space required when rasterizing scenes with high depth complexity. When low graphics memory necessities are of utmost importance, k-buffer will objectively be thought-about as the foremost most well-liked framework which advantageously ensures the proper depth order on a subset of all generated fragments. Although varied alternatives are introduced to partially or completely alleviate the noticeable quality artifacts produced by the initial k-buffer algorithm within the expense of memory increase or performance downgrade, applicable tools to automatically and dynamically compute the most appropriate worth of k are still missing. To the present finish, we introduce k+-buffer, a fast framework that accurately simulates the behavior of k-buffer during a single rendering pass. Two memory-bounded information structures: (i) the max-array and (ii) the max-heap are developed on the GPU to concurrently maintain the k-foremost fragments per pixel by exploring pixel synchronization and fragment culling. Memory-friendly strategies are further introduced to dynamically (a) reduce the wasteful memory allocation of individual pixels with low depth complexity frequencies, (b) minimize the allotted size of k-buffer per completely different application goals and hardware limitations via a simple depth histogram analysis and (c) manage local GPU cache with a fastened-memory depth-sorting mechanism. Finally, an intensive experimental evaluation is provided demonstrating the advantages of our work over all previous k-buffer variants in terms of memory usage, performance cost and image quality.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here