DART Investigation Log

6.30.2005

Song downloaded and installed DART 2.0 for Mac OS X.

7.5.2005

Tried to run DART test file LiveVideo. Unsuccessful.

7.6.2005

Tried to run DART tutorial 1.3 test file for camera test. Unsuccessful. Do we have the right type of camera. What type of camera do we have? We may have more sucess with DART when we get a fast PC.

7.11.2005

Successfully ran Dart test file MarkerTracking.dir using primitive objects. Solution was adding the right file path for data files and changing video size to 320x240. Unsuccessful in using shockwave 3d models instead though the objects would work when just dragged and dropped onto the timeline. Wrote an email to the Georgia Tech peeps for help on using Shockwave 3D objects instead of the primitives. Found that DART has the ability to use tracking methods other than simple fiducial marker tracking, including GPS, InertiaCube, and Intersense products. This would be a very big plus allowing the arena to be free of the clutter of cryptic visual markers everywhere one needed augmented content to appear.

7.12.2005

Got response on email inquiry. Tried resetting scale of objects and changing duration of actors on the timeline to no avail. Tried to create a Shockwave object that replicated the primitive in every way including size (6cm) and default textures but still got no display of the augmented object. Primitives work fine. Got an additional offer for help by someone willing to look at importing our Shockwave 3D object in place of the primitive. Waiting for reply. Song suggested her computer did not meet the minimum recommended requirements.

7.13.05

Still waiting for that reply. Making a short screen capture video to document DART success so far.

7.14.05

Caught up on our documentation. Looked at script and focused on one part of code regarding model hierarchies and display but didn't know where to go from there. Got response from Maribeth pointing out the solution in that area we had focused on. Figured out that the marker patterns used by DART are mapped to the same coordinate system, creating a mega-pattern.

7.15.05

Tried to get propeller/plane animation to appear unsuccessfully. Couldn't even get the plane to appear. Attempted rescaling the plane with no success. Decided to create a more simple test animation of a cube that would oscillate on x, y, and z axes. The cube appeared but didn't animate.

7.18.05

Song suggested looking at the LINGO behavior for playing keyframe animations. This led to the discovery of the "Actor Actions" DART event which allowed one to play an animation upon receiving a "cue". A simple time-based cue set to start playing the animation immediately allowed us to successfully see the cube/axes animation. Based on this success, we tried the more complex plane/propeller Shockwave 3D model animation and were also successful with this. However, in doing so we noted several artifacts with regard to the translation of models from Cinema 4D to Shockwave 3D format--deformations and array copy instances and some textures didn't appear to work.

7.19.05

Continuing from the previous day's success, we investigated how to initiate different kinds of cues to start, stop, and manipulation animations. Specifically, we were able to use the keyboard to control animation sequences. This may prove to be very useful since we may pass SMS type messages between Mission Control and the Window to control augmented content. We successfully placed two separate augmented objects upon the stage. Figured out how to standardize model scales. Discovered this by changing the size of a planet that our little Star Trek ship orbited without changing the scale of the ship. Exported models aren't given absolute units but are simply given relative numerical values, therefore having objects of the same scale is as simple as having the overall sizes be numerically close. Successfully associated augmented content with a different set of fiducial markers. We experimented somewhat to no avail with changing marker orientation definitions to learn more about them. Found a weird off-ratio tracking problem with some files but this was fixed by rebuilding the .dir file. We figured out how to standardize the augmented models' orientation and how the axes were translated upon import/export. This was related in part to figuring out the mapping and the world coordinate system for both the markers and how the objects were displayed in the augmented view.

7.20.05

Looked at DART Wizardry and Physics. Wizardry appears to allow communication between to different DART applications using standard IP addressing (cool). This may also lead to a technical solution to Mission Control/Hubble Window communications. The Physics behaviors would allow gravity and other aspects to be added to our 3D models with "relative" ease. Spent 4 hours trying to catch up on documentation but not getting much done due to video compression problems. Emailed Gavin about FTP access problem.

7.21.05

After giving up, we decided to live with out-of-sync audio in a Flash Video (FLV) format for video documentation of our successes. Caught up on remaining documentation and started to work on putting our site together and uploading it. Practically finished with DART investigation but would like to see how a DART application appears on the PC especially since it is likely that a mobile Hubble Window will be on a PC platform due to Apple's lack of such products.