Papers by Joseph Paradiso
My research at the MIT Media Laboratory has entailed close collaboration with many artists, techn... more My research at the MIT Media Laboratory has entailed close collaboration with many artists, technologists, and frequent interaction with managers at our corporate sponsor companies, affording a rare opportunity to work with and observe traditionally separate sub-cultures that do not often intersect. Contemporary design similarly inhabits a middle ground between art and engineering, as the designer's product should both work efficiently and make an aesthetic statement. Although the work of a pure artist isn't necessarily directed by an application, it's nonetheless generally inspired by some kind of stimulus in the real world that spurs the artist to express him or herself through their medium. It's fascinating to trace the process by which an artist conceives of and produces a piece. Often some sort of random-seeming stimulus acts as a seed that crystallizes and grows into a beautiful production. Although the final outcome of this process often bears little relation to the idea that launched the effort, the generation and evolution of the initial concept is a critical step in the artist's production. I've recently seen some striking examples of this process through a collaboration with the well-known kinetic artist Michael Moschen, who meticulously documents the various stages through which his performance pieces pass as they develop. He generally begins with a fascination with particular shapes or by noticing the way in which something moves in the environment, e.g., a branch flapping in the wind or the way a piece of litter rolls down the street. This inspires him to realize some sort of simple sculpture or aggregation of objects, which he often builds up in his garden. He watches how this structure looks and moves naturally in the environment, occasionally tweaking it himself, usually for weeks or more. Then he makes another set of objects that he can physically manipulate and thereupon begins to practice intensively with these artifacts to develop his performance. The process of design usually begins with a concept not necessarily generated by the designer. Unlike the artist, the designer is more often commissioned to produce something that serves a specific purpose. That said, designs that have the most impact involve breaking boundaries and seeing a particular function from a very different viewpoint. Accordingly, designers tend to create work environments that tap from and overlap various disciplines and approaches. I've had occasional opportunity to collaborate with several leading design firms and visit them at their workplaces. One …
We describe a 15 to 20 minute interactive live perform- ance work entitled Situated Trio for augm... more We describe a 15 to 20 minute interactive live perform- ance work entitled Situated Trio for augmented guitar and two computer musicians with expressive controllers. This work brings into focus a number of issues concern- ing musically expressive control and interaction among performers.
We present a novel approach to the creation of user-generated, documentary video using a distribu... more We present a novel approach to the creation of user-generated, documentary video using a distributed network of sensorenabled video cameras and wearable on-body sensor devices. The wearable sensors are used to identify the subjects in view of the camera system and label the captured video with real-time human-centric social and physical behavioral information. With these labels, massive amounts of continually recorded video can be browsed, searched, and automatically stitched into cohesive multimedia content. This system enables naturally occurring human behavior to drive and control a multimedia content creation system in order to create video output that is understandable, informative, and/or enjoyable to its human audience. The collected sensor data is further utilized to enhance the created multimedia content such as by using the data to edit and/or generate audio score, determine appropriate pacing of edits, and control the length and type of audio and video transitions directl...
IEEE Pervasive Computing, Jul 1, 2018
There is an opportunity now to shape the future of air quality monitoring using the lessons from ... more There is an opportunity now to shape the future of air quality monitoring using the lessons from other industries. We advocate for an inclusive, open, distributed ecosystem that unifies regulatory-grade data with that of more inexpensive devices. This system incentivizes transparency and good data practices, encourages engagement, and accelerates learning. It creates the opportunity for sophisticated, distributed air quality models; supports community-driven innovation; and gets actionable data in the hands of those who care. This paradigm is not just the future for air quality, but also the future for sensor ecosystems in general. Growing concern over air pollution has led to increases in air quality monitoring investment globally. In the US, more than 300 federal organizations collect air quality data, research funding has increased over the past several years, and citizen monitoring is on the rise. 1-3 With this increased focus, the overall market for air monitoring equipment is projected to grow from $3.4 billion in 2015 to at least $6 billion by 2022. 4 Unfortunately, this mounting engagement has not resulted in a commensurate increase in our understanding of air quality or an effective market. The diverse backgrounds and incentives that have driven growth in the air quality measurement ecosystem have also prevented effective collaboration among its instrument manufacturers, application developers, research scientists, technologists, entrepreneurs, citizens, and policymakers. The status quo is failing to mature and deliver useful insights, despite the revolution in sensor technology. We believe there is insufficient structure to catalyze a dynamic and rapidly evolving marketplace. The time is right for charting a vision forward as the private community grows, federal funding priorities shift, and affordable monitoring hardware becomes ubiquitous. An open and inclusive system based on communal norms would drive adoption, promote innovation, and
We share a study on the public adoption the Chibitronics circuit sticker toolkit, an open source,... more We share a study on the public adoption the Chibitronics circuit sticker toolkit, an open source, commercially available hardware toolkit for learning and creating electronics on paper. We examine sales data over a twoand-a-half-year period from November 2013, when the kit was launched commercially, to June 2016. We also look at publicly available project documentation from users during this period. We find that the Chibitronics user community confounds norms for traditional technology-making communities, especially in gender demographics. We explore the artifacts and types of documentation produced by users to learn about the various backgrounds, values, and goals of subcommunities, which includes educators, Makers, and crafters. In particular, we focus on artifacts from the craft community as a surprising and distinctive subset of technology creators. The diversity in public engagement shows how paper electronics tools like Chibitronics can be an effective approach for engaging new and broader audiences to participate in technology creation.
While wearable computers are empowering fashion accessories, they bring with them a tangle of war... more While wearable computers are empowering fashion accessories, they bring with them a tangle of wares which connect their parts. As these subsystems begin to decouple and operate on less power, it becomes possible to wirelessly distribute their required data and power using to the wearer's body. We have demonstrated systems that transmit and receive both data and power, and are working t o combine the two.
Scientific American, Jun 17, 2014
ABSTRACT How a world filled with sensors will change the way we see, hear, think and live
New Interfaces for Musical Expression, May 1, 2002
Passive RF Tagging can provide an attractive medium for development of free-gesture musical inter... more Passive RF Tagging can provide an attractive medium for development of free-gesture musical interfaces. This was initially explored in our Musical Trinkets installation, which used magnetically-coupled resonant LC circuits to identify and track the position of multiple objects in real-time. Manipulation of these objects in free space over a read coil triggered simple musical interactions. Musical Navigatrics builds upon this success with new more sensitive and stable sensing, multidimensional response, and vastly more intricate musical mappings that enable full musical exploration of free space through the dynamic use and control of arpeggiatiation and effects. The addition of basic sequencing abilities also allows for the building of complex, layered musical interactions in a uniquely easy and intuitive manner.
The MIT Faculty has made this article openly available. Please share how this access benefits you... more The MIT Faculty has made this article openly available. Please share how this access benefits you. Your story matters. Citation Alexander Reben and Joseph Paradiso. 2011. A mobile interactive robot for gathering structured social video. In Proceedings of the 19th ACM international conference on Multimedia (MM '11). ACM,
An increasing number of internet-connected LED lighting fixtures and bulbs have recently become a... more An increasing number of internet-connected LED lighting fixtures and bulbs have recently become available. This development, in combination with emerging hardware and software solutions for activity recognition, establish an infrastructure for context-aware lighting. Automated lighting control could potentially provide a better user experience, increased comfort, higher productivity, and energy savings compared to static uniform illumination. The first question that comes to mind when thinking about context-aware lighting is how to determine the relevant activities and contexts. Do we need different lighting for reading a magazine and reading a book, or maybe just different lighting for reading versus talking on the phone? How do we identify the relevant situations, and what are the preferred lighting settings? In this paper we present three steps we took to answer these questions and demonstrate them via an adaptive fivechannel solid-state lighting system with continuous contextual control. We implemented a multidimensional user interface for manual control as well as an autonomous solution using wearable sensors. We enable a simple set of sensors to manipulate complicated lighting scenarios by indirectly simplifying and reducing the complexity of the sensor-lighting control space using human-derived criteria. In a preliminary user study, we estimated significant energy savings of up to 52% and showed multiple future research directions, including behavioral feedback.
Lecture Notes in Computer Science, 2002
We describe an inertial gesture recognition framework composed of three parts. The first is a com... more We describe an inertial gesture recognition framework composed of three parts. The first is a compact, wireless six-axis inertial measurement unit to fully capture three-dimensional motion. The second, a gesture recognition algorithm, analyzes the data and categorizes it on an axis-by-axis basis as simple motions (straight line, twist, etc.) with magnitude and duration. The third allows an application designer to combine recognized gestures both concurrently and consecutively to create specific composite gestures can then be set to trigger output routines. This framework was created to enable application designers to use inertial sensors with a minimum of knowledge and effort. Sample implementations and future directions are discussed.
Physical Review Letters, Dec 24, 1979
Measurements of the reactions e++e e++e, p'+p, and 7 + T at PETBA energies (s = 13,17,27.4, 30, a... more Measurements of the reactions e++e e++e, p'+p, and 7 + T at PETBA energies (s = 13,17,27.4, 30, and 31.6 GeV) are reported. The results show that these reactions agree well with the predictions of quantum electrodynamics thus determining that all the known charged leptons are pointlike particles to a distance &-10~8 cm.
IEEE Pervasive Computing, Apr 1, 2020
This course is a broad introduction to a host of sensor technologies, illustrated by applications... more This course is a broad introduction to a host of sensor technologies, illustrated by applications drawn from human-computer interfaces and ubiquitous computing. After extensively reviewing electronics for sensor signal conditioning, the lectures cover the principles and operation of a variety of sensor architectures and modalities, including pressure, strain, displacement, proximity, thermal, electric and magnetic field, optical, acoustic, RF, inertial, and bioelectric. Simple sensor processing ...
Uploads
Papers by Joseph Paradiso