
This paper presents a novel real-time live compression scheme for 3D animated meshes in the context of client-GPU cloud gaming. We specialize in compressing character models animated with linear blend skinning (LBS) with rigid bones, using skinning decomposition to approximate skinning weights and bone transformations as closely as possible to the originals. Our technique leverages the Dem Bones library and proposes adjustments to handle common industrial constraints, such as low bone count decomposition of multicomponent meshes and reconstruction of uniformly scaled mesh poses. We demonstrate the efficiency of our real-time compression technique on meshes with up to 25,000 vertices and 200 bones, achieving compression rates that significantly reduce bandwidth usage while maintaining visually acceptable distortion levels (KGE ≤ 5%). Here, we show that our approach can effectively address edge cases encountered in video games, thanks to our optimizations and adjustments. Our work contributes to advancing the state of the art in graphics streaming for cloud gaming applications.
Real-time compression, Cloud gaming, Skinning decomposition, [INFO.INFO-GR] Computer Science [cs]/Graphics [cs.GR], Graphics streaming
Real-time compression, Cloud gaming, Skinning decomposition, [INFO.INFO-GR] Computer Science [cs]/Graphics [cs.GR], Graphics streaming
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 1 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
