Animated characters are frequently used in television programs, movies, and video games, but relatively little is known about how their characteristics affect attention and viewer opinions. We used eyetracking to examine the role of visual complexity, animation style, and motion complexity on viewing patterns and ratings of videorecorded and animated movie clips. We created twelve sets of videos using HD recording and animation methods. Each set consisted of four videos of an actor performing and describing a series of actions with blocks that resulted in a design. Of the videos, one was a regular HD recording of the actress (Video condition). The remaining videos were animated using motion capture (MoCap) data from that actress: (1) a 3d-rendered, photorealistic human character created to look like the actress (Virtual) animated using high-density facial MoCap data; (2) a 2d-rendered, cartoon-style human character that was created by toon-rendering the virtual character, using the same data (Cartoon); and (3) a 3d-rendered, photorealistic robot character animated using only a small subset of the facial motion capture data (Robot). This method ensured that motion characteristics were controlled while visual factors were varied. Participants answered questions about each video and the characters. There was a significant effect of character style on the percentage of video time spent viewing the face such that viewers looked least at the face of the robot followed by the video, cartoon, and virtual characters. Increased facial looking time correlated with unpleasantness and repulsiveness ratings. Overall ratings showed that the characters differed on how lifelike, likable, and creepy they were. In conclusion, animation styles have an effect on both viewing patterns and audience membersa?? subjective opinions of characters.
展开▼
