Blizzard patents automatic prop placement with machine learning

Video games have gained in visuals and immersion over the decades. The size of game maps has also seen a drastic increase over the decades, increasing the complexity of detailing a map along with the placement of props for developers. This is one particular reason why AAA video games take so long to develop these days.

We came across the recently released patent from Blizzard Entertainment that seeks to completely revolutionize the placement of props in a game using a trained AI system. This approach could be used in all sorts of titles, from AAA games to indie games, which require some sort of prop placement on a board.


Main takeaways:

  • Blizzard Entertainment recently released a patent that uses trained AI to automate the process of placing props on a map.
  • The patent discusses various computational methods and techniques to effectively implement this approach.
  • This innovative approach will significantly reduce development times and increase the immersion of game cards in the future.

The patent called “PROPS PLACEMENT WITH MACHINE LEARNING“elaborates on a prop placement tool that uses”a trained machine learning mechanism” to place props anywhere on a map. The tool will be formed”based on one or more training cards on which props have been placed.

Comparison of a map before and after prop placement using the prop placement tool.

Props can be anything that adds detail to a map like barrels, wooden boxes, and physically affected objects, to name a few. Placing these props manually turned out to be a laborious job for the developers, but the patent will significantly change the method of its implementation.

The figure shows various instances of target props to place on a target map using the prop placement tool.

The prop placement tool can use various advanced computational techniques or a combination of advanced computational techniques to dynamically implement props into a game map. Blizzard’s patent iterates, “the machine learning mechanism can be trained to suggest placement based on (a) relative spatial rules, (b) prop-specific rules, (c) distances between props and fixed objects between props and structures maps, and (d) distances between props.”

The figure shows the possible orientations of a specific accessory.

At the end of the training,the prop placement tool may be provided as input to (a) map data that defines a target map and (b) prop data that specifies the set of target props to be placed on the map target.“The prop placement tool can start generating a suggested location for each of the target props, on the map after calculating the input.

Blizzard’s patent dives into many factors that decide where a prop can be placed, one of the methods is “map locations.” “The “location” of a placement can be fine-grained (e.g. match a single pixel on the target map) or coarse-grained (e.g. match clusters of pixels or larger areas on the target map)….the number of possible slots for a target accessory is equal to the number of pixels on the target map.

Another variable that is considered is spatial rules. Blizzard’s patent talks about the “no collision rules.” Before applying the rule, each location is a possible candidate for the placement of a prop. However, the no-collision rule will make prop placement more natural and hassle-free.

The patent reads: “Based on a “collision-free” spatial rule that two objects cannot occupy the same space at the same time, candidate possibilities (location, orientation) for a target prop are selected to remove locations in which the periphery of the target prop would intersect/collide with the periphery of a map structure or with the periphery of an already placed target prop.

Prop-specific rules can also be applied to make prop placement feel natural to a game’s needs. Various rules include, but are not limited to:the prop must be placed with a particular edge of the prop against the edge of a card structure,” and “the prop must be placed next to a corner created by walls in the map structure”, among many others.

Many more technical details are discussed which can be read in Blizzard’s patent. The patent contemplates making a top view of a target board to apply proper calculations for prop placement. However, this approach can be very restrictive in some cases. The patent also clarifies a solution regarding this obstacle.

The figure shows how a two-dimensional image can be created from a target map, where the image is used to determine if locations satisfy one or more spatial rules.

The patent reads: “the two-dimensional top-down image of a map can indicate obstacles that should not limit object placement.

The patent gives an example of such a problem and also offers a viable solution. It is said, “the two-dimensional top-down image of a target map that is used for collision detection for a target object is taken at a particular height in the target map.”

The figure shows the distinction that a two-dimensional image taken from a top-down view of the entire board differs from a two-dimensional image taken from a top-down view of the target board taken at a height particular.

Blizzard Entertainment gives a perfect explanation of how this prop placement tool will revolutionize current development times. The patent reads: “For a virtual world that includes 5,000 scenes, the number of prop placements increases to 1,250,000.

Developers would have to spend a lot of time creating a AAA game due to the need to manually place props. However, prop placement can be automated using this machine learning tool.

Overall, this innovation will significantly reduce the time spent designing game cards and increase the visual appeal of future game cards without the manual labor required. AAA video games will see a reduction in development times by relying more on this intriguing technology.

What are your thoughts on automatically placing props on a map using AI technology? Let us know your thoughts in the comments below.

Related readings: Toshimichi Mori Leaves Arc System Works After 20 Years

Was this article helpful?

Good work! Thank you for giving your positive opinion ⚡

How could we improve this post? Please help us. ✍

Sherry J. Basler