DETAILS


COLUMNS


CONTRIBUTIONS


ARCHIVE



Web3D RoundUp: Looking Backwards and Forwards

Vol.34 No.2 May 2000
ACM SIGGRAPH


Remotely Operated Vehicle (ROV) Dive Visualization



Mike McCann
Monterey Bay Aquarium Research Institute (MBARI)



McCann, Figure 1: The virtual world of Monterey Bay with vertical exaggeration.

I was relieved when I saw the call for submissions to the Web3D VRML 2000 3DWeb RoundUP. In previous years I had been a reviewer and submitter of Web3D content, where all I had to work with was a URL. In most cases a website by itself cannot convey the full meaning and usability of rich 3D web content. This is especially true for 3D virtual worlds that are freely navigable. For the Web3D VRML 2000 Web3D RoundUP, a possible submission entry could be a one- to six-minute long videotape. This meant that I could have my submission efficiently explain the content and demonstrate its usability. Ironically, this low-tech medium (standard NTSC VHS videotape) was the best way to preview how I might present my work live to the Web3D RoundUP jury.

I made the videotape in the way I would present it to the audience. It had to be quick and to the point. I worked hard to get it down to four minutes in length. The video served its purpose. My work was accepted, and I was given three minutes for my live presentation. (Okay…I can shave off one more minute.)

During the testing period five hours before the show, everything worked great. The new GeForce 256 card in a 700 MHz system was giving me 17 frames per second, the wireless Internet connection was fast and everything was ready. I was feeling comfortable. About 10 minutes before my presentation I got onto the stage to check again that everything worked and discovered that my web server was not answering. This was not good!

Since this was being presented at the Web3D RoundUP, I had decided to actually use the web for my presentation. Besides, the essence of what I was presenting was the integration of 3D content into an on-line web database system. Also, it would have been difficult to have transferred the whole massive system to work on a stand-alone computer without a network connection. There was only one solution: I had to restart the web server at my home institution. At 9 p.m. there certainly wasn’t going to be anyone to call; the only way to do it was for me to gain dial-in access from my laptop which was the only computer around that had the special software needed to restart the server. There were no analog phone lines in the conference center so I begged a friend in the audience to let me use his hotel room phone. We ran there, I jacked in, restarted the server, and was back on the stage in 15 minutes.

I had been an audience participant for three previous Web3D RoundUPs and enjoyed giving immediate audience feedback with the noisemakers. Now it was my turn to be on the receiving end of all this noise. I hoped that the combination of stage fright and the concentration I needed to stay within the time limit would prevent me from hearing the noisemakers. This was not the case. My presentation involved about 45 seconds of surfing through traditional 2D web pages before getting into the 3D content. I deemed this necessary for giving the audience appropriate background to also demonstrate how the 3D content was integrated into a production web database application. Well, this was another tough audience…after the first 30 seconds I heard the first ‘moo’ (the “I don’t like it” noisemaker) and within 10 seconds there were at least 20 people mooing. I knew I had to speed it up and get into the 3D.

The mooing didn’t last long. I clicked on the 3D Replay link and within a few seconds I was navigating within a 3D virtual world of the submarine canyon in Monterey Bay. I zoomed into the submersible dive area and the audience could see the 3D models of the ship and ROV. This is when I heard the first squeak from the “I like it” noisemaker. By now the moos had stopped. Rotating around and showing how we’ve integrated real data, such as navigation track lines and color coding of dissolved oxygen onto the track line, brought more squeaks. I knew then that I had communicated what I had wanted. The last thing to do was to zoom way into the world and show an actual video frame grab of a transparent squid. The simulated noise of the ROV became louder along with the noise of more squeaks. I looked up at the countdown timer and there was one second left. It worked, it was over and the audience voted the ROV dive visualization as the best educational demo of the show.

This work can be accessed on-line at www.mbari.org/~mccann/vrml/ROVDataVis/.

Mike McCann
Monterey Bay Aquarium Research Institute (MBARI)



The copyright of articles and images printed remains with the author unless otherwise indicated.