At the end of the experiment, in order to evaluate the quality of
the obtained animation, we asked a panel of animation experts and
producers to assess if the quality was acceptable for the creation of
extra characters. The resulting scene, together with the production
process, was presented to a focus group of five animation producers,
video makers, and trainers, in a public panel. The aim of the focus
group was to assess the quality of the animation produced through
the reuse of action animations, the feasibility of the pipeline (including
the semantic labeling and the search and reuse phase), and to
validate the production scenarios for which the system was designed
(offline animation and animation of extra characters). The experts
evaluated its quality and potential for use in the animation
industry and the impacts of the technique for the animation languages.
According to the comments we collected, the system has a
high potential in the animation production, for both main and extra
characters, in the case of TV series productions, especially if coupled
with animation data extracted from motion capture takes. However,
it should be noticed that the major problems concern the creation of
the animation catalogue and the classification system that allows for
a fast retrieval of the animation segments. Past experiences in series
production that tried to create a repository of animation segments
lead to a waste of time in classification and retrieval. From the discussion,
it seems that an action-based storing and retrieval of animations
would facilitate the task of animators: in most past
experiences, producers reported about the frustration of animators
in retrieving their own work, with the consequence that most of
them tended to animate the same action again.