Skip to Content

3-D Design for the Masses

Dryad’s new user interface aims to make it easier for novices to create realistic objects for virtual worlds.
January 17, 2008

Recent years have seen an explosion in user-generated content of all types on the Web, but the explosion has been most intense where the content is relatively easy to produce. While users are welcome to design flora and artifacts for virtual worlds such as Second Life, modeling 3-D content is still fairly difficult for the inexperienced. A new technique called collaborative design-space exploration, developed by the Virtual Worlds Group at Stanford University, aims to make it easy for anyone to create 3-D designs for virtual worlds. The group has prototyped its interface in a program called Dryad, which allows users to design trees, and plans to extend it for use with other types of objects.

Digital arboretum: Dryad, a program that demonstrates a new user interface developed by the Stanford University Virtual Worlds Group, allows the user to design a 3-D virtual tree by exploring a map.

Vladlen Koltun, an assistant professor at Stanford who heads the group, says that he hopes design programs like Dryad will one day make it easy for anyone to create compelling content for virtual worlds, without having to learn a scripting language, or how to use a sophisticated 3-D modeling tool. He particularly hopes to make it easier for academics without computer-science expertise, for whom “content creation is one of the bottlenecks,” to put together virtual worlds for educational or experimental purposes.

Most 3-D modeling requires the user to start by modifying polygons and sticking them together, as if she were manipulating lumps of clay. When Dryad opens up, a user instead sees an overhead view of a broad selection of possible trees. The trees may look as if they were culled from forests, mountaintops, bayous, and alien planets. The user can then pan through the space or zoom in and out using an interactive map interface similar to that found in programs such as Google Maps. When the user sees a tree she likes, zooming in on it produces a new map with a new set of trees whose features resemble those of the one she selected. The closer the user zooms in, the more similar to each other the trees on the map will be. The user can click one of the visible trees to select it, or click a space between trees to see a new tree that blends the characteristics of surrounding trees. For finer control, the user can flip to a screen with “scrubber bars”–sliding switches that cover a continuous range of values–where she can adjust 100 different characteristics of the tree one at a time. While the scrubber bars are standard fare for 3-D design, the map is not. “The idea is, you always have something in front of you,” says Koltun, adding that inexperienced users are commonly intimidated by having to start from scratch. “The map interface guides you to regions of the space that you might have overlooked, and shows you ideas of what you can make.” The interface allows users to design a near-infinite variety of trees, Koltun says.

Although the interface appears simple, there’s a lot of sophisticated math behind it. In order to make the user feel as though she is exploring the space of possible designs, Koltun says, the software has to represent the 100 different parameters governing the appearance of the trees on two-dimensional maps. Those 100 parameters translate mathematically into a space of 100 dimensions. Koltun explains that every choice the user makes causes the program to choose a curved two-dimensional slice of the true space, in a process similar to that used to draw a line of best fit for a set of points in a scatterplot. Naturally, some information is lost in the process, but since the slice curves through multiple dimensions, it allows the user to make changes to far more than two parameters at once.

The researchers also designed the program to intelligently decide which possible trees to display to users. “Showing a set of random designs would be awful,” Koltun says, explaining that choosing random values for the 100 possible parameters would likely result in distorted, unappealing trees. So in determining which trees to display, the program is guided by the selections of previous users.

Dryad isn’t the only way for users to create 3-D designs. Second Life has a scripting language that can produce anything from a tree to a Zamboni, but it’s a language that the user has to learn. Interactive Data Visualization’s SpeedTree is a sophisticated tree-design tool that players of Unreal Tournament 3 can use to design trees for customized gaming environments. But company president Michael Sechrest explains that the tool is geared to players with an interest in game development, called modders, and it hasn’t been adjusted for ease of use.

Peter Phillips, technical director for Millions of Us, an agency that specializes in creating content for virtual worlds, considers Dryad “a really promising tool to allow inexperienced people to create personalized content.” He adds that “one of the things that looks unnatural in MMOs [massively multiplayer online worlds] is when we see repeated identical objects.” Phillips thinks that tools such as Dryad can solve that problem by making it easy for people to tailor objects to their personal tastes. While he considers the map interface a good approach, he says that the current version of Dryad has some performance shortcomings.

Right now, Dryad is available for Windows and Macintosh systems, and the Stanford group is working on a Linux version. Koltun notes that Dryad is only a demonstration of a type of design interface, and he says that the group plans to extend the concept to allow users to design human forms and other objects commonly found in virtual worlds.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.