=Paper=
{{Paper
|id=None
|storemode=property
|title=Design of a Perceptual-based Object Group Selection Technique
|pdfUrl=https://ceur-ws.org/Vol-588/123.pdf
|volume=Vol-588
}}
==Design of a Perceptual-based Object Group Selection Technique==
https://ceur-ws.org/Vol-588/123.pdf
Design of a Perceptual-based Object Group Selection Technique
Hoda Dehmeshki∗ Wolfgang Stuerzlinger†
York University
A BSTRACT and anchor edge e, the algorithm finds all paths of length four start-
Selecting groups of objects is a common task in graphical user in- ing from 1 and along edge e, see also figure 1. We call these paths
terfaces. Current selection techniques such as lasso and rectangle primary paths. Then, for each primary path, the method computes a
selection become time-consuming and error-prone in dense con- linearity coefficient (LC) that indicates how strongly the four nodes
figurations or when the area covered by targets is large or hard to are perceived as a line:
reach. This paper presents a new pen-based interaction technique !
that allows users to efficiently select perceptual groups formed by (a21 + a22 − 2ra1 a2 )
LC = exp − ,
the Gestalt principle of good continuity. 2s2 (1 − r2 )
1 I NTRODUCTION where a1 and a2 are the angles between lines connecting the center
Selecting groups of objects is a common task in graphical user of objects, and r and s are experimentally determined constants [2].
interfaces and is required for many standard operations. Current Primary paths with a LC smaller than a threshold are discarded
selection techniques such as lasso and rectangle selection become as they are unlikely to be perceptually salient.If at least one primary
time-consuming and error-prone in dense configurations or when path remains, the algorithm continues as follows: for each path it
the area covered by targets is large or hard to reach. Perceptual- searches its potential continuations by identifying all neighbors of
based selection techniques considerably reduce selection time when the last object in the path. Then, the four-object window is shifted
targets form perceptual groups, as predicted by Gestalt principles to include each neighbor and the LC is computed in turn. If the
of proximity and good continuity. However, they use heuristic and new LC is smaller than a threshold, that completion is ignored, oth-
not-validated grouping algorithms. Also, they do not allow editing erwise, the extension of the path with this neighbor is added to a
of a selection or selecting of groups with random configurations. stack. If all of the neighbors of the last node are unacceptable, the
Dehmeshki and Stuerzlinger developed a perceptual-based object original path is kept; otherwise, the original path is discarded since
group selection technique for mouse-based user interfaces [1]. In at least one good extension has been found. The algorithm contin-
their system double-clicking on an object that is part of multiple ues until all paths inside the stack have been processed.
(curvi-)linear groups selects all the groups. To deselect an unde- Figure 1 illustrates this. In Fig. 1-a, the anchor node ‘1’ and an-
sired group, the user alt-clicks on its first non-desired object. Three chor edge ‘e’ are represented by thick borders. All four-vertex win-
key elements distinguish that system from the present work. First, dows with small LC are visualized with ovals, while the ones with
clicks can specify only the location of a group but not the direction large LC are shown in rectangles. There are three primary paths,
in which a group of objects extends. This makes selection less ef- see Fig. 1-b, but only the one inside the rectangle is considered for
ficient when objects belong to multiple groups. Second, it provides further extension. After two more iterations, see Fig. 1-c and d,
no support for selecting non-perceptual groups. Finally, their sys- only the straight path is returned as being perceptually plausible,
tem relies heavily on multiple-clicks, which is not appropriate for see Fig. 1-e.
pen-based systems. This problem is shared by other techniques that A B C
use multi-clicking to cycle through different perceptual interpreta- 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
tions [3]. e
2 P ER S EL D E
1 2 3 4 5 6 1 2 3 4 6
5
This paper introduces PerSel, a new pen-based object group selec-
tion technique, which addresses the mentioned problems. PerSel
consists of two components: The first component detects good con-
tinuation groups based on a neighborhood graph. The second pro- Figure 1: Illustration of Good continuity grouping algorithm. (a)
vides a set of pen-based interaction techniques that use the detected shows the anchor object 1 and edge e. (b) the rectangle and ovals
groups to facilitate path-based selection. visualize primary paths (c) the path inside the rectangle is extended.
(d) shows new potential primary paths. the one inside the rectangle
2.1 Detecting Good Continuation Groups is extended. (e) objects 1-6 are grouped.
The system first constructs a neighborhood graph. When the user
performs a straight flick gesture starting from inside an object, the For line gestures that cross an object, the gesture is first decom-
system examines all edges in the neighborhood graph that are con- posed into two half-gestures, using the closest point of the gesture
nected to this object and picks the one which has the closest direc- relative to the center of the object. Then, the linear groups corre-
tion and distance to the gesture. The object and the edge are called sponding to each half is found using the above methods, and then
the anchor object and the anchor edge, respectively. the two groups are merged.
PerSel is based on an implementation of Feldman’s model [2] for Curvilinearity group detection: Similar to the linear case,
linear groups which models paths as groupings of four objects com- when the user performs an arc gesture starting from inside an ob-
bined with a sliding window paradigm. Given an anchor object 1 ject, the system examines the edges in the neighborhood graph that
are connected to this object. The edge closest to the gesture is
∗ e-mail: hoda@cs.yorku.ca picked, as defined by the sum of distances between gesture points
† e-mail: wolfgang@cs.yorku.ca and the edge. The rest of algorithm works similar to the linear case,
33
except that: (1) we use the curvilinearity coefficient that adapts the
O1 O1 O1
formula for LC by using the deviation from the average angle in-
stead of the angles themselves, and (2) in the initial phase we only O2 O2 O2
consider primary paths that turn in the same direction as the gesture.
2.2 Gestural Interaction A A A
In this section we explain the gestural interaction techniques avail-
able in PerSel. As common in pen-based systems, tapping on a
single object selects it. Also, PerSel cancels all selections when-
A B C
ever the user taps the pen on the background. This affords a simple
and fast way to cancel erroneous selections.
Figure 5: Resolving ambiguity. (a) An arc gesture on object O1 se-
Path Selection: Performing a straight gesture across an object lects both curvilinear groups. (b) A “cutting” gesture disambiguates
selects the Good Continuation group aligned with the gesture direc- which objects to select. (c) Only desired objects remain selected.
tion. Similarly, an arc gesture across an object selects the curvilin-
ear group that has a similar direction as the gesture. In both cases,
the selected group is visualized by links connecting the successive
enable selection of such paths, we introduce a new path editing fea-
objects, see Fig. 2.
ture. Assume that a path is already selected. If the user draws a
gesture across an already selected node (called an anchor), a sup-
plementary anchor is created. Then the selected path is modified by
(1) automatically deselecting objects on the path beyond the new
anchor, and (2) adding the new (curvi-)linear group corresponding
to the new gesture and anchor to the selection, see Fig. 6.
(a) (b) (c) (d) O1
Figure 2: Good continuity group selection. Performing a line (a) or
O2 O3
an arc (c) gesture, selects the corresponding linear (b) or curvilinear
(c) group and visualizes it by links.
Partial Path Selection: There are two alternatives for partial (a) (b) (c) (d)
selection of paths. The first way is to select the complete path and
then cut undesired part(s) by drawing a flick gesture across one (or Figure 6: (a) Target objects have thick borders. (b) A line gesture
two) visualized links, see Fig. 3. The second alternative is to initiate over O1 selects the corresponding group. (c) A gesture from O2
the selection by a flick gesture from inside an object, see Fig. 4. guides the selection. (d) A gesture from O3 adds the remainder of
the desired objects.
(a)
(b) 3 C ONCLUSION AND F UTURE W ORK
This paper presented PerSel, a new gesture-based selection tech-
(c) nique that is based on the Gestalt principle of Good Continuation.
Performing a flick gesture crossing an object selects the (curvi-
Figure 3: Partial group selection. (a) A flick gesture across an object )linear group(s) that the object belongs to and is aligned with the
selects the whole group. (b) Two flick gestures deselect all objects gesture direction. PerSel also provides interaction techniques that
beyond these “cut” gestures. allow users to perform partial group selection and selecting groups
with arbitrary configurations. As future work we will include
Gestalt principle of similarity and extend PerSel to deal with ob-
jects with different visual features such as shape and size.
R EFERENCES
[1] H. Dehmeshki and W. Stuerzlinger. Intelligent mouse-based object
group selection. In Smart Graphics, 2008.
[2] J. Feldman. Curvelinearity, covariance, and regularity in perceptual
(a) (b) (c) (d)
groups. Vision Research, 37(20):2835–2848, 97.
[3] E. Saund and T. Moran. A perceptually-supported sketch editor. In
Figure 4: Partial group selection. In (a) respectively (c) a line and an Proceedings of the ACM Symposium on User Interface Software and
arc gesture start from inside an object. Only objects that are on the Technology–UIST’94, pages 175–184, New York, 1994. ACM.
same side of gesture are selected, as in (b) and (d).
Resolving Non-Perceptual Groups If a gesture corresponds to
multiple potential curvilinear groups, all of them are selected. The
user can then disambiguate the section by deselecting the non-
desired groups. This is similar to the partial selection technique,
in that the paradigm of “cutting” links is used to separate the non-
desired objects from the targets, see Fig. 5.
Selecting Paths With Multiple Segments: More complex paths
often consist of connected Good Continuity groups (segments). To
34