Outdoor Mixed Reality

figure summary


If a user wants to see an object hidden by a blocking object, the see-though vision system displays the invisible object by seeing through the the blocking object.In the see-through vision with a tablet device, motion parallax is the most important depth clue because of its high portability.It has not been clarified that motion parallax by moving the tablet device affects user's perception.We classified the motion parallax on tablet-based see-through vision and analyzed its performance on user's depth perception.

Shun Kawahi,Itaru Kitahara,Yoshinari Kameda,and Yuichi Ohta,"Depth Perception Analysis by Motion Parallax in See-Through Vision with Tablet Device ,"IEICE Technical Report, vol.114, no.487, MVE2014-80, pp.33-38 (2015.3)

hayashi http://www.image.esys.tsukuba.ac.jp/wp-content/uploads/2014/02/hayashi.png 574w" sizes="(max-width: 300px) 100vw
300px" />

Diorama-based interface that displays a point of interest (POI) on a miniature of real 3D world is a good approach to share the POI with people working in outdoor environment. The viewpoint to observe the diorama model should be manipulated by users to explore the diorama model and find the POI. However, poorly designed viewpoint manipulation method may cause difficulty to understand the corresponding point of the POI in the real world. A viewpoint manipulation method should be able to manipulate the viewpoint freely and the viewpoint enables a user to understand the correspondence between the real world and the diorama model easily. In order to realize a viewpoint manipulation method with satisfying the above requirements, this paper compares three viewpoint manipulation methods (free shot, pan/tilt shot and dolly-around/crane shot) that utilize a mobile device pose. We have implemented an AR (Augmented Reality) test bench of the diorama-based interface with photorealistic diorama model to conduct the subjective evaluation experiment. As a result, we found that dolly-around/crane shot is superior to the others in aspect of performance to find POI and subjective impressions.


Masayuki Hayashi,Itaru Kitahara,Yoshinari Kameda,and Yuichi Ohta,“A User Study on Viewpoint Manipulation Methods for Diorama-Based Interface Utilizing Mobile Device Pose in Outdoor Environment,“ 21st International Conference on Artificial Reality and Telexistence (ICAT2011),pp.41-46,2011.

nagashima http://www.image.esys.tsukuba.ac.jp/wp-content/uploads/2014/02/nagashima.png 674w" sizes="(max-width: 300px) 100vw
300px" />

This paper proposes a new Mixed-Reality presentation of spatio-temporal behaviour of a moving object in a observer viewpoint. The behaviour of the moving object was recorded in advance as a sequence of pairs of spatial position and its texture by environmental camera. A reviewer can see simultaneous replays of the moving object so that he/she can shorten the time to browse the behaviour. Since large number of simultaneous replays increases the chance of overlapping textures, we propose the method to determine the best number of replays. We also propose an enriched visualization of trajectory of the moving object so that the reviewer can recognize the moving direction of the object easily. We have conducted subject tests on our proposed traject visibility and simultaneous replays of the moving object and show the results.

長島正典,林将之,北原格,亀田能成,大田友一,録画された移動物体行動の複合現実型提示における一覧性の向上; 日本バーチャルリアリティ学会論文誌,17,4,pp.447-456 (2012.12)