imp table

an interactive installation by rebecca bray

The Imp Table is an interactive installation that allows users to communicate with creatures made of shadow. If users choose to engage with the imps and follow the narrative that unfolds, they must make a choice between saving or abandoning the imps that they have created.

The imp table was born out of a desire to create compelling interactive stories that jump out of the screen to create real illusion and immersion in a changed world. While the table seems initially merely charming, the experience demands that the user confront their own desire for or repulsion against engaging in a world that is more complicated than it first may seem. The project incorporates video, programmatic and frame-based drawn animation, projection, and video sensing.

View storyboard  |  Read background

background

Prototype 1: The first prototype of the project was developed by creating storyboards, taking video of myself as the imp, processing those images, and created frame-by-frame animation in flash. I then mapped those animations to my physical layout with a table projected upon from above. The intention was to have proximity sensors under the table set off one of 3 animations. The user touched the plate and an imp emerged, waved at the user, and proceeded to put shadows of peas and a chicken leg on the plate. He then jumped from the table and went to sleep. If the user touched him when sleeping, the imp became angry.

See the video above of the first prototype, or view documentation of the animation and physical computing process. This phase also incorporated motors that pushed the plate and cup up when the imps emerged from beneath them.

In user testing of the first prototype, the feedback was overwhelmingly positive in terms of reaction to the movement and nature of the animation. Users loved to watch and engage with the imps, and expressed delight at the strange juxtaposition of 2 and 3-D elements. The first prototype also demonstrated that more complex interactive tools were necessary to create an experience that furthered that delight beyond initial charm.

Prototype 2: Because proximity sensors only offer on-off sensing, I began to explore video sensing as a way to allow users to interact more fully with the imps. I also moved beyond frame-by-frame animation into Actionscript-based programmatic animation that allows further complexity.

Future development: After exploring the potential of video sensing and programmatic animation for this piece, I plan on deepening the narrative as part of a larger project of examining how to create compelling interactive narrative in installation.

prototype 1 development process (completed) more details here

Interface Elements:

• The table itself, constructed so that magnets and motors could be used to manipulate objects.
• Objects on the table, constructed from scratch or modified. Painted white.

PComp physical elements:

• Sensors to react when users touch objects on the table.
• Motor and magnet system to move
• Sensors to chip, chip to laptop to action script, laptop to projector, laptop to motor/magnets

Computing Elements:

• Action Script - Data from touch sensors to trigger projection of flash animation

Animation:

• Flash-based animation with Action Script

Challenges:

• Getting microprocessor to send data to Flash
• Lining up animations with physical objects
• Motors and magnets - getting these to work well with animation.
• Touch sensors sensitive enough to work through large objects, without being way too expensive.

Process:

• Research other similar projects
• Research how to get chip to send data to action script
• Decide on content and specific intent
• Buy sensors and test them
• Start with demo animation
• Build prototype with sensors and motors and do user testing
• Research for content
• Finalize plan
• Get touch sensor to action script part working
• Get laptop out to motor working
• Create final animation
• Build final table and object pieces
• Testing
• Final presentation, documentation

Initial Animation Research:

Character Development.

To get a sense of the angle and movement of the Imps.

The second phase of development was begun at the beginning of 2006, after the first prototype had been tested and filmed.

• Evaluate sensors for best possible complex reaction scenerios
• Explore video sensing, do testing
• Modify animations for Actionscript programmatic movement
• Create a series of storyboards and narrative possibilities for stages of project
• Write Actionscript code to test with video sensing
• Get video sensing working with table setup, try infrared light
• Write code based on storyboard to create variety of interactions
• Speed up processing by using streaming technology on flvs for animations
• Increase complexity of interactions, including proximity and direction, in java

prototype 3 development process (intended)

• Reevaluate java vs jitter for video sensing, rewrite code and testing
• Further develop narrative tree
• Shoot video of new scenes
• More animation and video processing
• Testing, rebuilding, testing...