17-03-2014, 10:09 PM
ABSTRACT
A virtual hand can intentionally deviate from the real hand
configuration to improve interaction and experience. Subsequent
discrepancy reduction (convergence) is important to avoid
undesirable side effects such as grasping difficulty. Prior work did
not study convergence motion effects adequately to understand
finger motions and speed effects. We present the first graspingfocused
studies on convergence motion speed effects. The results
are important for optimizing convergence motion and can
contribute to guidelines for related interaction types.
Keywords: Virtual grasping, visual-proprioceptive discrepancy.
Index Terms: H.5.1 [Information Interfaces and Presentation]:
Multimedia Information Systems—Artificial, augmented, and
virtual realities
1 INTRODUCTION
We previously showed benefits of separating a virtual hand from
the real hand configuration. Preventing visual hand-object
interpenetration improves subjective grasping experience [1], and
maintaining resulting joint angle offsets to lift fingers at release
improves performance [2]. Other motivations include improved
hand navigation [3] and pseudohaptic effects [4].
Subsequent virtual-real discrepancy reduction (convergence
that removes the offset) is important to avoid artifacts, including
unreasonable virtual hand configuration [2]. Researchers suggest
convergence algorithms (Section 2) but did not study convergence
motion effects adequately to understand finger motions and speed
effects. Very slow speed maintains discrepancy for too long and
causes grasp problems [2]. Fast speed can negate discrepancy
benefits so that virtual fingers stick to object surfaces during
release in case of substantial real finger penetration [2]. Visualproprioceptive
motion discrepancy [3] may affect user experience
as converging virtual finger motion differs from real motion.
Understanding convergence effects is important for optimizing
interactions, e.g., finding convergence speeds for good grasp and
release performance or user-preferred finger motions. We present
a grasping-focused study on convergence motion effects and
finger convergence speed. It investigates performance effects of
speed, a possible relationship to user motion speeds, and a
subjective user adjustment of the related motion discrepancy.
A virtual hand can intentionally deviate from the real hand
configuration to improve interaction and experience. Subsequent
discrepancy reduction (convergence) is important to avoid
undesirable side effects such as grasping difficulty. Prior work did
not study convergence motion effects adequately to understand
finger motions and speed effects. We present the first graspingfocused
studies on convergence motion speed effects. The results
are important for optimizing convergence motion and can
contribute to guidelines for related interaction types.
Keywords: Virtual grasping, visual-proprioceptive discrepancy.
Index Terms: H.5.1 [Information Interfaces and Presentation]:
Multimedia Information Systems—Artificial, augmented, and
virtual realities
1 INTRODUCTION
We previously showed benefits of separating a virtual hand from
the real hand configuration. Preventing visual hand-object
interpenetration improves subjective grasping experience [1], and
maintaining resulting joint angle offsets to lift fingers at release
improves performance [2]. Other motivations include improved
hand navigation [3] and pseudohaptic effects [4].
Subsequent virtual-real discrepancy reduction (convergence
that removes the offset) is important to avoid artifacts, including
unreasonable virtual hand configuration [2]. Researchers suggest
convergence algorithms (Section 2) but did not study convergence
motion effects adequately to understand finger motions and speed
effects. Very slow speed maintains discrepancy for too long and
causes grasp problems [2]. Fast speed can negate discrepancy
benefits so that virtual fingers stick to object surfaces during
release in case of substantial real finger penetration [2]. Visualproprioceptive
motion discrepancy [3] may affect user experience
as converging virtual finger motion differs from real motion.
Understanding convergence effects is important for optimizing
interactions, e.g., finding convergence speeds for good grasp and
release performance or user-preferred finger motions. We present
a grasping-focused study on convergence motion effects and
finger convergence speed. It investigates performance effects of
speed, a possible relationship to user motion speeds, and a
subjective user adjustment of the related motion discrepancy.