Skinned Mesh Animation Of The Programmable Hardware-based Design And Realization

Character animation is among the most active fields in Computer Science in the recent decade, it is widely applied in Virtual Reality, CAD, Advertising and Digital Entertainment, etc. As computer hardware, especially consuming-level video cards with built-in acceleration capability technology progresses, Real-time Character Animation is becoming more and more widely used; it is of a great economic prospect and theoretical research significances. This thesis makes deep research into Programmable Graphics Hardware and Real-time Character Animation and presents a way to implement the most widely used Real-time Character Animation--Skinned Mesh Animation in Programmable Graphics Hardware.There are three types of Real-time Character Animation: Articulation Animation, Single Mesh Model Animation and Skinned Mesh Animation. This thesis describes all three with considerable depth. Skinned Mesh Animation can be seen as the combination of Articulation Animation and Single Mesh Model Animation, it has the advantages of the former two while lacking their shortcomings, but Vertex Blending Technique used in Method Skinned Mesh Animation requires a lot of matrix computations, these used to be performed in the CPU, which will incur a great burden on CPU and reduce the real-time quality of animation. With Programmable Graphics Hardware now we have a possible solution to this problem.This thesis describes Programmable Graphics Hardware and its programming language and method in great detail. Nowadays the growing speed of computing capability of GPU of graphics hardware has surpassed that of CPU. A breakthrough in graphics hardware technology is the introduction of programmable function, which allows developers to replace the original fixed pipeline function modules with their own customized Shader programs to achieve more flexibility. Although GPU has a very high computing speed, algorithms implemented in CPU cannot be put to execute in GPU directly because of the discrepancy in instruction execution manner of the two. GPU's architecture is a high parallel SIMD instruction set system. To reimplement algorithms insufficient to run on CPU with Programmable Graphics Hardware, it has to reconsider the data structures and procedures to implement them to make full use of the...