Vis Comput (2017) 33:1357–1369
DOI 10.1007/s00371-016-1231-2
ORIGINAL ARTICLE
Kinetic depth images: flexible generation of depth perception
Sujal Bista
1
· Ícaro Lins Leitão da Cunha
2
· Amitabh Varshney
1
Published online: 6 May 2016
© Springer-Verlag Berlin Heidelberg 2016
Abstract In this paper we present a systematic approach to
create smoothly varying images from a pair of photographs
to facilitate enhanced awareness of the depth structure of a
given scene. Since our system does not rely on sophisticated
display technologies such as stereoscopy or auto-stereoscopy
for depth awareness, it (a) is inexpensive and widely acces-
sible, (b) does not suffer from vergence - accommodation
fatigue, and (c) works entirely with monocular depth cues.
Our approach enhances the depth awareness by optimizing
across a number of features such as depth perception, opti-
cal flow, saliency, centrality, and disocclusion artifacts. We
report the results of user studies that examine the relationship
between depth perception, relative velocity, spatial perspec-
tive effects, and the positioning of the pivot point and use
them when generating kinetic-depth images. We also present
a novel depth re-mapping method guided by perceptual rela-
tionships based on the results of our user study. We validate
our system by presenting a user study that compares the out-
put quality of our proposed method against other existing
alternatives on a wide range of images.
Keywords Kinetic depth effect · Depth perception ·
Aesthetics in visualization · Visualization for the masses
B Sujal Bista
sujal@umiacs.umd.edu
1
Department of Computer Science and Institute for Advanced
Computer Studies, University of Maryland, College Park,
USA
2
Universidade Federal Rural de Pernambuco - Unidade
Academica de Garanhuns, Garanhuns, PE, Brazil
1 Introduction
The kinetic-depth effect (KDE) is the perception of the
three-dimensional structure of a scene resulting from a rotat-
ing motion. First defined by Wallach and O’Connell [53],
the kinetic-depth effect has been used widely. Images that
exhibit KDE are commonly found online. These images
are variously called Wiggle images, Piku-Piku, Flip images,
animated stereo, and GIF 3D. We prefer to use the term
Kinetic Depth Images (KDI), as they use the KDE to give the
perception of depth. In 2012, Gizmodo organized an online
competition to reward the KDI submission that best provided
a sense of depth [55]. Recently, the New York Public Library
published a collection of animated 3D images online [18].
Flickr has a large number of groups that discuss and post
animated 3D stereo images. Also, online community-based
galleries that focus on KDI can be found at the Start3D
website [8]. With the increasing availability of stereo and
lightfield cameras (such as the Lytro), the use of the KDE to
express depth on ordinary displays is rising rapidly.
Although the basic form of the KDE is trivial to imple-
ment for a virtual environment where the 3D geometry and
lighting are known, excessive motion and reduced depth
perception mar the visual experience unless proper care is
taken. In fact, accomplishing a high-quality KDE from a
pair of photographs is not trivial and has several interesting
nuances, as outlined in this paper, which should be useful for
any practitioner wishing to use this technique for facilitating
depth awareness. In this paper we describe an algorithm and
its accompanying system, for facilitating depth awareness
through KDE using only a pair of photographs or images.
The use of the KDE is a viable alternative to the use of
stereoscopic and autostereoscopic displays: (i) the KDE pro-
vides monocular depth cues that allow us to experience depth,
even with one eye, which accommodates people who suffer
123