Jun Rekimoto, Squama: Modular Visibility Control of Walls and Windows for Programmable Physical Architectures, AVI 2012

7/31/2019 Jun Rekimoto, Squama: Modular Visibility Control of Walls and Windows for Programmable Physical Architectures,

1/4

Squama: Modular Visibility Control of Walls and Windowsfor Programmable Physical Architectures

Jun RekimotoInterfaculty Initiative in Information Studies,The University of Tokyo

7-3-1 Hongo, Bunkyo-ku, Tokyo, JapanSony Computer Science Laboratories

3-14-13 Higashigotanda, Shinagawa-ku, Tokyo

[email protected]

ABSTRACT

In this paper we present Squama, a programmable physi-cal window or wall that can independently control the vis-ibility of its elemental small square tiles. This is an ex-ample of programmable physical architecture, our vision forfuture architectures where the physical features of architec-tural elements and facades can be dynamically changed andreprogrammed according to peoples needs. When Squamais used as a wall, it dynamically controls the transparencythrough its surface, and simultaneously satisfies the needsfor openness and privacy. It can also control the amountof sunlight and create shadows, called programmable shad-ows, in order to afford indoor comfort without completelyblocking the outer view. In this paper, we discuss how infuture, architectural space can become dynamically change-able and introduce the Squama system as an initial instancefor exemplifying this concept.

Keywords

architectural space, programmable matter, programmablearchitecture, interaction design

Categories and Subject Descriptors

H.5.2 [Information Interfaces and Presentation]: UserInterfacesInteraction styles, Prototyping

General Terms

Design, Human Factors

1. INTRODUCTIONTraditional physical architectures, such as buildings and

houses, are designed to protect people from external threats,such as chills, heat, wind, noise, and enemies. To achieve thisgoal, physical architectures have to be hard, solid, inflexible,

Permission to make digital or hard copies of all or part of this work forpersonal or classroom use is granted without fee provided that copies arenot made or distributed for profit or commercial advantage and that copiesbear this notice and the full citation on the first page. To copy otherwise, torepublish, to post on servers or to redistribute to lists, requires prior specificpermission and/or a fee. AVI 12, May 21-25, 2012, Capri Island, ItalyCopyright 2012 ACM 978-1-4503-1287-5/12/05 ...$10.00.

and static. Moreover, once constructed, it is difficult tochange its shape and function. However, this inflexibilityalso prohibits peoples dynamically changing demands forarchitectural spaces.

On the other hand, computer software is flexible, scal-able, and adaptable. It can be programmed to accommo-

date users changing needs. Biological elements, such as thechameleons skin, can also be considered as an another ex-ample of such flexibility. It not just covers the chameleonsbody (like a wall), but its visual pattern also changes byresponding to the outer environment. In a similar manner,human skin also slowly responds to sunlight by increasing ordecreasing its melanin production. We consider that thesedynamic and flexible features are also desirable for futuristicphysical architectures.

Recently, the research concepts such as ProgrammableMatter [7], Claytronics [5], Organic User Interfaces [8], Pro-grammable Reality [6], or Radical Atoms have emerged to in-corporate the flexibility of software into physical things. Forexample, a group of small particle robots can cooperatively

and dynamically construct a physical structure. Hence, webelieve that physical architecture should also become moreflexible, programmable, and reconfigurable, and we call sucha concept programmable physical architecture.

Many elements constitute physical architectural space. Asan initial step for programmable architectures, we focus onthe wall and the window because both are fundamental ele-ments. In a physical environment, walls are used to separateand shield spaces. On the other hand, windows are used forallowing outer light into the room. Although the functionsof windows and walls are related, people often have con-tradicting demands from them. For example, satisfying theprivacy and openness needs at the same time is not alwayspossible.

This paper presents Squama, a programmable wall andwindow that can change its transparency and opacity inmodular way. Squama is composed of a grid of small liq-uid crystal (LC) tiles, called phygcells and each phygcellstransparency can be independently controlled. It can be-come a wall (when all the LC tiles are opaque) or a window(when all the LC tiles are transparent). It can also be pro-grammed to act as an intermediate between a wall and awindow. Squama can be used as a wall that can separaterooms or cubicles. In addition, it can be used as a windowto the outer space.

168


2/4

Figure 1: Squama: modular programmable windowor wall for dynamic architecture. Each square panel(called phygcell) can independently change its trans-parency.

2. SQUAMA: MODULAR TRANSPARENCY

CONTROL FOR ORGANIC ARCHITEC-

TURE

Squama is a flat surface that is designed as a window, apartition in the physical architecture (Figure 1). It controlstransparency in a modular way. As a bitmap screen consistsof a 2D array of computer-controlled pixels, Squama consistsof a 2D array of small LC tiles (that we call phygcells)that can control its transparency. Although our currentlyimplemented Squama prototype has coarse phygcells (i.e.,10cm10cm in size and the grid size is 188), it is possibleto increase the granularity of arrays in order to make thesystem more similar to standard computer displays.

Unlike that of other wall-sized displays, Squamas mainpurpose is to control rays that pass through the surface.Although it can be used as a low-resolution bitmap display,it is more interesting to use it to create a programmable

shadow, and simultaneously satisfy the openness and privacyneeds.

In the following three sections, we explain three typicalapplications of Squama to demonstrate how it can be usedin physical environments.

2.1 Real-World PixelizationLack of privacy is often a problem for people living in an

urban area. People living in a condominium may be dis-turbed by people living in nearby buildings. Moreover, un-necessary eye contact with nearby people can be very annoy-ing. Hence, to completely preserve privacy, a quick solution

Figure 2: Real-world pixelization: real-world viewthough the Squama can be selectively controlled:(above) hiding confidential video content, and (bot-tom) eliminating unnecessary eye contacts.

would be to obscure the visibility of the window by using athick curtain or blinds; however, it also hampers the venti-lation of the home and blocks the natural scenic view fromthe window. In an office, preserving openness and maintain-ing privacy are also often incompatible demands. Meetingrooms separated by a solid wall hamper openness; however,a transparent glass wall may not be able to protect privacyand secure information.

View control is another unsatisfied demand. If the windowcould selectively show views that people like (e.g., trees andmountains) and hides views that people do not like (e.g.,nearby buildings), the view from the window would becomemore comfortable. However, just statically hiding a part ofthe window does not serve the purpose, because the area tobe hidden changes depending on the viewers position.

In summary, transparency (e.g, openness) and opacity(e.g., privacy and security) are contradicting requirementsfor physical windows or walls. Squama solves this problemby acting as a modular transparency control surface. Asa typical graphics display controls each pixel, the Squamawindow consists of a grid of small units, and each opaquearea can be dynamically selected by a computer.

Our current implementation of Squama is 180cm 80cmand divided by 188 square modules. Each modules trans-parency is switchable by the control command from the sys-tem. Using this configuration, this window can selectivelyhide a view. This is similar to image pixelization that is usedfor masking information on an image. In normal pixeliza-tion, a part of the image is blurred for displaying it in lowerresolution. In our case, a part of a natural view is obscuredby controlling the transparency of the units (Figure 2). Wecall this real-world pixelization.

One important feature of real-world pixelization isviewerdependence. While normal image pixelization is appliedto two-dimensional images, the masking area can be deter-

169


3/4

Figure 3: Programmable Shadows: when a user de-fines an area that must be protected from the directsunlight, Squama automatically determines the areaof the window to make a proper shadow.

mined only by the content of the image. In our case, theposition of the person in front of the Squama window isconsidered. When the person moves, the area to be maskedalso changes.

To handle this situation, our system measures peoplesviewing positions. We use Microsoft Kinect and its SDK [2]for people tracking. Kinect is a depth sensing camera basedon structured projection, and it can obtain an image withdepth information. Kinect SDK provides skeletal track-ing where the skeletal information of people in front of theKinect sensor can be estimated. We extract the persons 3Dhead position, and then change the masking area accordingto this information.

We tested two application cases. In the first case, thetarget object is static, and only the viewer moves. This caseis for protecting information in a room, or hiding objectsyou do not want to see. In the second case, both the targetobject and the viewer move. This case is for preventing eyecontract between people over the window.

2.2 Programmable ShadowsAnother use of Squama is to control the amount of sun-

light and create a shadow. Although one of the main roleof the window is to let the natural light enter the room, un-necessary light becomes a problem. For example, ultravioletrays in sunlight can damage things in the house. We oftenencounter such a situation at a restaurant or at a cafe; thesunshine directly shines on the table and people, makingthem very uncomfortable. Many people avoid such a tableor try to avoid the rays by manually closing the curtain orblind.

In our solution, people can define things that must beprotected from sunlight. Then, the system computes thedirection of the sun and automatically calculates the partof the window area that must be shield. In other words,our system lets users control shadows instead of manuallycontrolled curtains or other sun-shielding methods. We callthis programmable shadows (Figure 3).

In our current implementation, we provide a simple userinterface in order to let a user to define shadow areas. Then,the system identifies the area that should be protected bycomputing the direction of the sun. As the sun moves, thecorresponding area on the window to create a shadow alsochanges, and the created shadow stays on the target initiallydefined by the user.

Figure 4: Ambient Displays

We are also plan to install this mechanism to eliminatedirect sun reflection on the TV or computer display b ecauseit makes watching TV or display quite annoying. We expectthat such dynamic control of sunlight is beneficial for main-taining a reasonable temperature; thus,it reduces the energyconsumption of the house.

We can also define people (especially, children or babies)that must be protected from sunlight. In this case, we againuse Kinect for person tracking, Squama controls the shadowregion in order to always keep people within the created

shadows.In these cases, Squama no longer acts as an information

control device. It directly controls the physical resource (thesunlight). This feature is very different from standard com-puter displays.

2.3 Ambient DisplaysAlternatively, a window can serve as an ambient infor-

mation display [14]. The previous programmable shadowsexample, where the main purpose was to control sunlight,can also be used as a communication method. People be-come aware when a part of or the entire window flashes,as they notice shadow change caused by this effect. In thiscase, no explicit image pixels are used; however, people still

can obtain information by just controlling the transparencyor opacity of the window pixel. We believe this to be anew (and more calm [14]) way of providing information topeople.

The other way to inform is to control tile modules as pix-els. Although the pixel resolution of current implementationis not sufficient, it shows the potential of displaying infor-mation on a window (Figure 4).

3. SYSTEM CONFIGURATIONWe use a 10cm 10cm square switchable light control

LC panel UMU glass manufactured by NSG UMU Prod-ucts [3]. This LC panel is a thin LC filmaAAlaminatedwith two transparent electrodes. It is normally opaque (lighttransmission ratio is 6%). When a voltage of around 100Vis applied to the electrodes, it becomes transparent (lighttransmission ratio is 86%). The switching speed is about1/100s.

We laid out this module in a 18 8 grid on a transparentacrylic board. The size of the total system is 184cm90cm.To test various usage patterns, we hang this unit in severalplaces in an office and a home.

Each LC panel is independently controlled by the micro-processor. We used six Arduino Mega [10] broads for con-trolling 144 LC panels. These Arduino boards are connectedto a Squama server on a host computer, written in Pro-

170


4/4

cessing [11], and it receives panel modification commandsthrough the network.

4. RELATED WORKOur approach is inspired by smart glasses where the

transparency of interior partition glasses can be changed.Boeing 787 also introduced transparency-controllable win-dows. While those systems simply control the intensity of

the entire glass, our system provides more precise controlwith a modular transparency control design that satisfiesopenness and view control simultaneously.

The Squama can be regarded as an example of surfacecomputing [4] systems, where people interact with comput-erized tabletop- or wall-shaped surfaces. Our system aimsto control light that passes through the surface, rather thanto interact with image pixels on the surface. A notable fea-ture that differentiates our device is that it not only acts asa display, but also controls real-world resources (in our case,sunlight and shadows).

Izadi et al. created a tabletop system that uses an LCshutter-glasses [9]. When the projection light and LC shut-ter glasses are synchronized, images appear both on the

tabletop and a handheld LC board. BYU-BYU-View is asystem that transmits physical air-flow between two sur-faces [12]. It consists of an air-flow sensor and an arrayof fans. Communication between these surfaces can be en-hanced by controlling airblow. This work also showed thepotential of physical resources within physical architecturalspace.

FogScreen generates a screen in air (by a controlled airflowand mist), and show images with projection [1]. While it alsoshows the potential of dynamically created walls or windows,it is mainly used for a display, and precise transparency con-trol is not the purpose. Creating weightless walls [13] is an-other attempt to dynamically control sound space interac-tively. Using this system, people can spontaneously define

sound walls for shielding outer noise and maintaining pri-vacy. Currently this is rather a conceptual system; it is im-plemented by requiring all the users to wear noise-cancelledheadsets, and physical sound is not actually controlled.

Recently, with the advance in brightness of the projectionsystem, several researchers and media artists tried to di-rectly show images on the surface of existing architectures;this technique is called projection mapping. A careful cal-ibration of the physical architectures geometry and projec-tion images can create an impression that the real walls andwindows dynamically move. While this is still a visual il-lusion, our ultimate goal is to create such an architecture,where a real building can change its structure according tothe peoples needs. Starting from a completely transpar-

ent building, people would be able to define any part of thesurface to be opaque.

5. CONCLUSIONIn this paper we propose the concept of programmable

physical architecture, where the elements of physical archi-tecture become programmable and dynamically change theirphysical features in real time, according to the users de-mands and environment. As a first example of this concept,we created Squama, a wall or window whose transparencycan be controlled in a modular way. It satisfies the open-ness and privacy needs that are usually incompatible. It

can also control the amount of sunlight and create shadows(programmable shadows) to afford indoor comfort withoutcompletely blocking the outer view.

As an initial prototyping of our concept, the system hascoarse modules, and we are currently constructing the nextgeneration system that uses an LCD panel obtained from aTV unit.

6. ACKNOWLEDGEMENTSThe author wishes to express his appreciation to the ef-

forts made by members in the construction of the system. Inparticular, he extends a note of thanks to Yoshiki Takeoka,Kensaku Kawauchi, Michihiko Ueno and Wataru Yamada.

7. REFERENCES[1] FogScreen. www.fogscreen.com.

[2] Kinect for Windows SDK beta.research.microsoft.com/en-us/um/redmond/projects/kinectsdk.

[3] NSG UMU Products. www.umupro.com.

[4] ITS 10: ACM International Conference onInteractive Tabletops and Surfaces, ACM (2010).

[5] Aksak, B., Bhat, P. S., Campbell, J., DeRosa, M.,Funiak, S., Gibbons, P. B., Goldstein, S. C., Guestrin,C., Gupta, A., Helfrich, C., Hoburg, J., Kirby, B.,Kuffner, J., Lee, P., Mowry, T. C., Pillai, P. S.,Ravichandran, R., Rister, B. D., Seshan, S., Sitti, M.,and Yu, H. Claytronics: highly scalablecommunications, sensing, and actuation networks.SenSys 05, ACM (2005), 299299.

[6] Coelho, M., Poupyrev, I., Sadi, S., Vertegaal, R.,Berzowska, J., Buechley, L., Maes, P., and Oxman, N.Programming reality: from transitive materials toorganic user interfaces. CHI EA 09, ACM (2009),47594762.

[7] Goldstein, S. C., Campbell, J. D., and Mowry, T. C.Programmable matter. IEEE Computer 38, 6 (June2005), 99101.

[8] Holman, D., and Vertegaal, R. Organic user interfaces:designing computers in any way, shape, or form.Commun. ACM 51 (June 2008), 4855.

[9] Izadi, S., Hodges, S., Taylor, S., Rosenfeld, D., Villar,N., Butler, A., and Westhues, J. Going beyond thedisplay: a surface technology with an electronicallyswitchable diffuser. UIST 08, ACM (2008), 269278.

[10] Mellis, D. A., Banzi, M., Cautielles, D., and Igoe, T.Arduino: An open electronics prototyping platform. InCHI 07 Extended Abstracts, ACM (2007).

[11] Reas, C., and Fry, B. Processing: A Programming

Handbook for Visual Designers and Artists. MITPress, 2007.

[12] Sawada, E., Ida, S., Awaji, T., Morishita, K., Aruga,T., Takeichi, R., Fujii, T., Kimura, H., Nakamura, T.,Furukawa, M., Shimizu, N., Tokiwa, T., Nii, H.,Sugimoto, M., and Inami, M. BYU-BYU-View: a windcommunication interface. In ACM SIGGRAPH 2007emerging technologies, SIGGRAPH 07, ACM (2007).

[13] Takeuchi, Y. Weightless walls and the future office.CHI 10, ACM (2010), 619628.

[14] Weiser, M., and Brown, J. S. Designing calmtechnology. POWERGRID JOURNAL 1 (1996).

171

Documents

Jun Rekimoto, Squama: Modular Visibility Control of Walls and Windows for Programmable Physical Architectures, AVI 2012