A data compression method for impulse response deformation model

As an approach to real-time simulation of haptic interaction with deformable objects, we proposed a record-reproduction method called dasiaimpulse response deformation modelpsila (IRDM), in which the deformation of an object shape is computed by convolving temporal sequence of interaction force with...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2009 World Haptics Conference s. 428 - 433
Hlavní autoři: Tagawa, K., Hirota, K., Hirose, M.
Médium: Konferenční příspěvek
Jazyk:angličtina
japonština
Vydáno: IEEE 01.01.2009
Témata:
Boundary conditions
Character generation
Computational efficiency
Computational modeling
Computer interfaces
Computer simulation
Data compression
Deformable models
H.5.1 [Information Interfaces and Presentation]: Multimedia Information System-Animations
H.5.2 [Information Interfaces and Presentation]: User Interfaces-Haptic I/O
Haptic interfaces
High performance computing
I.6.5 [Simulation and Modeling]: Model Development-Modeling Methodologies
ISBN:9781424438587, 1424438586
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:As an approach to real-time simulation of haptic interaction with deformable objects, we proposed a record-reproduction method called dasiaimpulse response deformation modelpsila (IRDM), in which the deformation of an object shape is computed by convolving temporal sequence of interaction force with a pre-computed response of model to impulse force. Although, IRDM has an advantage that its computation cost of interaction force is independent of model complexity, it requires a huge amount of memory to store impulse response data corresponding to the combination of all input and output degrees of freedom. This paper discusses an approach to solve this problem by compressing the impulse response data. The key idea is to substitute similar impulse response waveforms by modification of a smaller number of representative waveforms. In our prototype implementation, a correlation coefficient was used as an index of the similarity. By this proposed approach the experimental implementation performance proved that the amount of storage was reduced to 1.1% of the original data size in cases where the threshold correlation coefficient value was 0.8.
ISBN:9781424438587
1424438586
DOI:10.1109/WHC.2009.4810859