Random Hole Display Left 90

Novel Displays

Ye Pan, Will Step­toe and Antho­ny Steed

Random Hole Display

We intro­duce a mul­ti-view autostereo­scop­ic telep­res­ence dis­play and its asso­ci­at­ed view-depend­ed ray traced ren­der­ing meth­ods. Our new sys­tem is inspired by the “Ran­dom Hole Dis­play” design  that mod­i­fied the pat­tern of open­ings in a bar­ri­er mount­ed in front of a flat pan­el dis­play from thin slits to a dense pat­tern of tiny, pseu­do-ran­dom­ly placed holes. This allows view­ers any­where in front of the dis­play to see a dif­fer­ent sub­set of the display’s native pix­els through the ran­dom-hole screen.

Multiview Display

Gaze, atten­tion, and eye con­tact are impor­tant aspects of face to face com­mu­ni­ca­tion, but some sub­tleties can be lost in video­con­fer­enc­ing because par­tic­i­pants look at a sin­gle pla­nar image of the remote user. We pro­pose a low-cost cylin­dri­cal video con­fer­enc­ing sys­tem that pre­serves gaze direc­tion by pro­vid­ing per­spec­tive-cor­rect images for mul­ti­ple view­points around a con­fer­ence table. We accom­plish this by using an array of cam­eras to cap­ture a remote per­son, and an array of pro­jec­tors to present the cam­era images onto a cylin­dri­cal screen. The cylin­dri­cal screen reflects each image to a nar­row view­ing zone. The use of such a sit­u­at­ed dis­play allows par­tic­i­pants to see the remote per­son from mul­ti­ple view­ing direc­tions. We com­pare our sys­tem to three alter­na­tive dis­play con­fig­u­ra­tions. We demon­strate the effec­tive­ness of our sys­tem by show­ing it allows mul­ti­ple par­tic­i­pants to simul­ta­ne­ous­ly tell where the remote per­son is plac­ing their gaze.

A Gaze-preserving Situated Multiview Telepresence System

Ye Pan and Antho­ny Steed

 @inproceedings{pan2014gaze,
  title={A gaze-preserving situated multiview telepresence system},
  author={Pan, Ye and Steed, Anthony},
  booktitle={Proceedings of the 32nd annual ACM conference on Human factors in computing systems},
  pages={2173--2176},
  year={2014},
  organization={ACM}
} 

Sphere Avatar Display

We report on two exper­i­ments that inves­ti­gate the influ­ence of dis­play type and view­ing angle on how peo­ple place their trust dur­ing avatar-medi­at­ed inter­ac­tion. By mon­i­tor­ing advice seek­ing behav­ior, our first exper­i­ment demon­strates that if par­tic­i­pants observe an avatar at an oblique view­ing angle on a flat dis­play, they are less able to dis­crim­i­nate between expert and non-expert advice than if they observe the avatar face-on. We then intro­duce a nov­el spher­i­cal dis­play and a ray-traced ren­der­ing tech­nique that can dis­play an avatar that can be seen cor­rect­ly from any view­ing direc­tion. We expect that a spher­i­cal dis­play has advan­tages over a flat dis­play because it bet­ter sup­ports non-ver­bal cues, par­tic­u­lar­ly gaze direc­tion, since it presents a clear and undis­tort­ed view­ing aspect at all angles. Our sec­ond exper­i­ment com­pares the spher­i­cal dis­play to a flat dis­play. Whilst par­tic­i­pants can dis­crim­i­nate expert advice regard­less of dis­play, a neg­a­tive bias towards the flat screen emerges at oblique view­ing angles. This result empha­sizes the abil­i­ty of the spher­i­cal dis­play to be viewed qual­i­ta­tive­ly sim­i­lar­ly from all angles. Togeth­er the exper­i­ments demon­strate how trust can be altered depend­ing on how one views the avatar.

Comparing Flat and Spherical Displays in a Trust Scenario in Avatar-Mediated Interaction

Ye Pan, William Step­toe and Antho­ny Steed

 @inproceedings{pan2014comparing,
  title={Comparing flat and spherical displays in a trust scenario in avatar-mediated interaction},
  author={Pan, Ye and Steptoe, William and Steed, Anthony},
  booktitle={Proceedings of the 32nd annual ACM conference on Human factors in computing systems},
  pages={1397--1406},
  year={2014},
  organization={ACM}
} 

Sphere Video Display

The move­ment of human gaze is very impor­tant in a face to face con­ver­sa­tion. Some of the qual­i­ty of that move­ment is lost in video con­fer­enc­ing because the par­tic­i­pants look at a sin­gle pla­nar image of the remote per­son. We use an array of cam­eras to cap­ture a remote user, and then dis­play video of that per­son on a spher­i­cal dis­play. We com­pare the spher­i­cal dis­play to a face to face set­ting and a pla­nar dis­play. We demon­strate the effec­tive­ness of the cam­era array and spher­i­cal dis­play sys­tem in that it allows observers to accu­rate­ly judge where the remote user is plac­ing their gaze.

Preserving Gaze Direction in Teleconferencing using a Camera Array and a Spherical Display

Ye Pan and Antho­ny Steed

 @inproceedings{pan2012preserving,
  title={Preserving gaze direction in teleconferencing using a camera array and a spherical display},
  author={Pan, Ye and Steed, Anthony},
  booktitle={3DTV-Conference: The True Vision-Capture, Transmission and Display of 3D Video (3DTV-CON), 2012},
  pages={1--4},
  year={2012},
  organization={IEEE}
}