Benko H , Holz C , Sinclair M , et al. NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld Virtual Reality Controllers[C]// the 29th Annual Symposium. ACM, 2016.

Abstract

We present an investigation of mechanically-actuated handheld  controllers  that  render  the  shape  of  virtual  objects through physical shape displacement, enabling users to feel3D surfaces, textures, and forces that match the visual rendering. We demonstrate two such controllers, NormalTouchand TextureTouch. Both controllers are tracked with 6 DOF and produce spatially-registeredhaptic feedback to a user’s finger. NormalTouch haptically renders object surfacesand provides force feedback using a tiltable and extrudable platform. TextureTouch renders the shape of virtual objects including detailed surface structurethrough a 4×4 matrix of actuated  pins.  By  moving  our  controllers  around  in  space while keeping their finger on the actuated platform, users obtain the impression of a much larger 3D shape by cognitively integrating output sensations over time. Our evaluation compares the effectiveness of our controllers with the two defacto standards in Virtual Reality controllers: device vibration and visual feedback only. We find that haptic feedback significantly increases the accuracy of VR interaction, most effectively by rendering high-fidelity shape output as in the case  of  our  controllers.  Participants  also  generally  found NormalTouch and TextureTouch realistic in conveying the sense of touch for a variety of 3D objects.

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

Contributions 
Our paper makes the following four specific contributions: 
1.  NormalTouch, a handheld controller that renders haptics through an active tiltable and extrudable platform and senses force input from the user upon touch. 
2.  TextureTouch, a handheld controller that renders the 3D surface structures via a 4×4array of actuated pins. 
3.  The integration of shape controllers in a VR system as well  as  a  series  of  solutions  to  interaction  challenges, 
such as object penetration and dynamic object behavior. 
4.  A user study comparing our two controllers with a visual-only and a vibrotactile feedback baseline, showing 
gains in accuracy and fidelity of haptic feedback.

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

 

 

Interfaces

Two devices shown in Figure 13 were used to complete the 
tasks  and  we  tested  four  interfaces:  NormalTouch,  TextureTouch, VibroTactileand VisualOnly

The VibroTactile interface activated the vibration motor whenever the bottom of 
a participant’s finger was within +/- 2mm of the surface, an  experience similar to haptic vibration feedback in game controllers 

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

In VisualOnly, the system provided no haptic output or additional visual feedback of any kind, beyond seeing their finger  penetrate  the  surface  of  the  object.

 

Tasks

Targeting Accuracy Task:During this task, participants were 
repeatedly presented with one of the three target types shown 
in Figure 15.

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

Tracing Accuracy Task: During  this  task,  participants  followed a 3D path with their fingertip as accurately as possible.
Tracing  paths  included  two  straight  lines  and  two  curved  lines, once appearing flat in front of the participant and once
rotated at a 35° angle (Figure 16).

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

Fidelity Assessment Task: During  this  task,  participants  pressed the clicker, saw each of the high-quality 3D models  shown in Figure 17, and explored their shape and surfaces  through touch by moving around the controller in the areas  marked in red.

Hypotheses 
We  hypothesized  that  haptic  feedback  would  increase  the 
level of fidelity of perceiving virtual objects, resulting in: 
H1.  Haptic  feedback  leads  to  more  accurate  targeting  and tracing compared to VisualOnlyfeedback
H2. NormalTouchand TextureTouch allow  targeting  with higher accuracy than VibroTactile, because they render 3D shapes with higher fidelity, facilitating precise touch. 
H3. TextureTouch produces the lowest error overall, because it renders structure on the participant’s finger as opposed to 
just the surface normal. 
H4.  Participants  complete  trials  fastest  in  the Visual Only condition, because no cues other than visual need cognitive 
attention and time to process

Results

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

UIST '16 paper NormalTouch and TextureTouch: High-fidelity 3D Haptic Shape Rendering on Handheld

Discussion

讨论很详细以后写论文可以参考

相关文章:

  • 2021-08-17
  • 2021-12-10
  • 2021-05-11
  • 2021-11-02
  • 2021-07-07
  • 2021-11-12
  • 2021-08-24
  • 2021-07-26
猜你喜欢
  • 2021-08-05
  • 2022-12-23
  • 2021-04-16
  • 2021-08-07
  • 2021-08-25
  • 2021-06-13
相关资源
相似解决方案