From CS 294-10 Visualization Sp10

Jump to: navigation, search

Lecture on Apr 19, 2010




  • Animation: Can it facilitate? Tversky, Morrison and Betrancourt. (pdf)
  • Principles of Traditional Animation Applied to Computer Animation. Lasseter (acm)

Optional Readings

  • Representing motion in a static image: Constraints and parallels in art, science, and popular culture. Cutting. (pdf)
  • On creating animated presentations. Zongker and Salesin. (pdf)

Ryan Greenberg - Apr 20, 2010 02:19:44 pm

As we discussed in class there is some disconnect between animation as Lasseter discusses it (how animation techniques can convey story, intention, emotion) and animation in visualization. Some of Lasseter's technical points seem useful, like the use of squash and stretch to avoid strobing. I although thought these two quotes were relevant:

"If a lot of action is happening at once, the eye does not know where to look and the main idea of the action will be 'upstaged' and overlooked."

This makes me think of visualization libraries with layout tools that animate the new position of nodes when they are rearranged. In many cases almost every element onscreen is moving, which makes it hard to track any of them. Is the main use of animation here to prevent jarring transitions?

"The animator must go to the heart of anything or any idea and develop its essence, understanding the reason for it, so that the audience will also understand it."

This resonated with me because it parallels the creation of visualizations for communication (as opposed to exploration). If we as creators don't understand the essence of what a visualization is supposed to communicate, we'll have a hard time making it lucid.

Also, I had always seen the larger Luxo lamp as a maternal figure--it was surprising to read Lasseter refer to it as "Dad."

Stephen Chu - Apr 26, 2010 01:48:52 am

Tversky: The paper cites some examples of when instructional animations did not significantly improve students' understanding when compared to static images paired with text. Tversky notes that one reason for this might be because animations can be comprehended discretely. I think instructional animations would be generally much better to use than static images when the motion taught should not be viewed discretely. One general example is learning a sports technique that requires fluidity. Instructional animations has been used heavily in sports because body position matters at all times during the techniques. Another example.. teaching someone how to perform a potentially dangerous weight lift should use animation because the motion should not be viewed step by step. The weightlifter needs to always be aware of his position throughout the lift, so the entire action should be captured in the graphic.

Jaeyoung Choi - Apr 26, 2010 03:14:30 am

Tversky: Many cases in this paper that show the failures of animation to benefit were quite interesting and actually I agree to some of them. Byrne(1999) reports that there is no benefit of animated graphics when teaching computer algorithms. I try to learn and understand algorithms by creating my own animation of the process as the algorithm goes. These recreating process of my own version in the brain really works better than just reading text and viewing sequence of static example. If shown with the animation, I may get stuck with the image it shows and have difficulty at recreating my own in the future.

This reminds me of what people say about books and their movie versions. (although books and their movie versions are far from conveying equivalent information as directors cut many things from book or add their own stuffs to the movies). Films leave little to viewer's imaginations. Readers of the book create their own movie in a sense, and decide how the characters speak, what they look like, and so on. This process of imagining and interpreting is what makes reading different from viewing a film.

As people have different capability and speed at perceiving information, adding interactivity to allow stopping, starting and replaying (to turn the animation into static image if needed) or zooming in/out and such may be what would make the animation facilitate.

DavidZats - Apr 26, 2010 10:12:05 am

In this lecture, we saw how powerful animation is as a tool. Not only can animation be used to attract the viewer's attention, it also prompts us to create stories based upon them motion. I found the most interesting aspects of this lecture to be about the challenges in properly using motion. Take for example an animation depicting the movement of parts within an engine. We see that it is difficult to distinguish between cause and effect. Additionally, all of the moving parts makes it harder to focus on any one element.

Arpad Kovacs - Apr 26, 2010 12:45:22 pm

Animation by Tversky:

The article evaluates whether animation facilitates better teaching of complex mechanical, biological, computational systems as static graphics. The paper cites and analyzes numerous research examples, and states that most research or their findings in this subject are skewed, because animation usually presents either additional information (for example, continuous animation depicts lower level actions as opposed to static graphics), or adds other elements (such as interactivity or prediction) which confound the results. Therefore, the author concludes that in the absence of controlled research that uses strictly the same information for animation and static graphic, previous findings voicing the usefulness of animation are baseless.

In my opinion, comparing static graphics and animation is like comparing apples and oranges. Since animation adds movement and timing to the picture, it is impossible to fairly compare to a static graphic, such as a set of small multiples showing the each step. There are too many variables that can influence the effectiveness of the animation (such as frame rate, overall duration of the animation, and selected pauses for emphasis) and I predict that there is also some individual variance over the most effective combination of these variables.

Principles of Traditional Animation by Lasseter:

I found the article about the 11 Fundamental principles of traditional animation very informative. By describing and explaining every principle, and illustrating them with pictures from animated movies, it provides a very useful guide for novices creating or at least understanding animations. The comments regarding applicability to computer graphics are a bonus. I especially liked the example on timing, describing how in-between drawings can completely change the dynamics and meaning of an action.

Jon Barron - Apr 26, 2010 05:49:44 pm


It's great that psychology journals allow people to publish negative results. I can't imagine a paper like this would fly at a graphics conference.


This is certainly an enjoyable paper to read. It's primary contribution doesn't seem to be its contents (as these ideas have been well-known in 2D animation, presumably) but the fact that it's tailoring these ideas to a 3D animation crowd. The only material that actually seems directed specifically towards 3D animation, and not 2D animation, is in relating "slow-in and out" to cubic spline interpolation.

Zev Winkelman - May 02, 2010 10:30:42 pm

Cutting - I had never considered breaking down methods of depictions of motion before, but I thought the paper did a decent job. However, the dynamic balance/stroboscopic/affine shear/forward lean/blur/action lines typology does not seem exhaustive. Figure 1 was an intense image.

Tversky, Morrison and Betrancourt - I sure hope animation can facilitate, because I'm trying to incorporate some of these ideas into a few of my projects. I will now have two more guiding principles - the congruence and apprehension principles. The paper mentions the lack of comparability between animations and static images which I had noticed in other works we have read. When the comparison is made, often times more than just the medium has been altered. Occasionally new information is coded or existing information is emphasized, complicating the comparison. I also liked the example of the galloping horse as an animation that was difficult to perceive before stop gap photography. And the point that sometimes discrete information is better conveyed in discrete images, as opposed to continuous animations

Lasseter - I loved this article. Maybe it was because I always wondered why that lamp was jumping around the screen in the opening of pixar animated movies. I had never considered how difficult it was for an animator to bring such an object to life. The principles of traditional animation explained many of these challenged.

Zongker and Salesin - These guys had me at "Our own lives would be improved if we could give - and receive - better talks." I think the decision to invest in creating a presentation using a scripting language would depend on how important the presentation was. The cost of learning the scripting language vs the benefits the scripting provided. The modeling concept seems pretty natural, but again, if I just need three pictures of flywheels and pulleys, maybe I'll just draw them. Unless I REALLY need the interactivity portion. Why not just use full blown graphics / animation packages and sprinkle in text when necessary ? I also thought it was interesting that these guys thought that the Luxo Jr. character animation techniques were distracting for the 'informing' that goes on during a presentation.

Prahalika Reddy - May 12, 2010 08:19:28 am

I found this lecture to be very enjoyable. Animation is one of the more interesting topics to study. I feel animations can be very effective in almost every case. Animating transitions is one of the most important uses of animations. It's very helpful to directly see how something changed. Oftentimes, it's very hard to follow what happened unless the transition has been animated. In the example we saw in class (with the shapes), I couldn't tell what had happened without the animation.

I also found it interesting that we were able to interpret a story from the animation with the two triangles and the circle. It was amusing to watch, but intriguing to see us all understand what was going on with no explanation other than the animation itself.

Causality is another interesting topic. The motion of the two balls clearly demonstrated this topic and it's very obvious that having the right timing is very important in effective animations.

Shimul Sachdeva - May 13, 2010 09:26:29 am

Stack, the demo showed in class - was appealing and yet hard to comprehend. Although animations are easy when it comes to narrating a story, they do not allow the user to view all the information at the same time. The lecture sort of blurred the divide between "interactivity" in visualizations and pure animation. For instance, we studied the Name Voyager demo under the topic of interactivity as well as animation. I guess the actual screen dynamics is animation and the controls that allow it are interactivity tools.

Studies on a human brain's cache capacities was fun. The moving airplane example was slightly confusing and the combustion engine animation was a good example. I have worked a little with Maya (animation/graphics software program) and the design principles discussed seemed to fit well with Maya's tools.

Lasseter: Luxo Jr.'s hop was an interesting example to demonstrate the principles of animation. One that I thought was missing though is zoom/focus during the animation. I think it is important enough to receive attention because it is sometimes best to focus into a particular aspect of the object, e.g., when only a part of the screen is in motion.

[add comment]
Personal tools