ARAnimator: In-situ Character Animation in Mobile AR with User-defined Motion Gestures

摘要

Creating animated virtual AR characters closely interacting with real environmentsis interesting but difficult. Existing systems adopt video seethroughapproaches to indirectly control a virtual character in mobile AR,making close interaction with real environments not intuitive. In this workwe use an AR-enabled mobile device to directly control the position andmotion of a virtual character situated in a real environment. We conducttwo guessability studies to elicit user-defined motions of a virtual characterinteracting with real environments, and a set of user-defined motion gesturesdescribing specific character motions.We found that an SVM-based learningapproach achieves reasonably high accuracy for gesture classification fromthe motion data of a mobile device. We present ARAnimator, which allowsnovice and casual animation users to directly represent a virtual characterby an AR-enabled mobile phone and control its animation in AR scenesusing motion gestures of the device, followed by animation preview andinteractive editing through a video see-through interface. Our experimental results show that with ARAnimator, users are able to easily create in-situcharacter animations closely interacting with different real environments.