利用者:Aligorith/GSoC2011 Proposal 2NextGenAnimTools

2. Next-Generation Animation Editing + Creation Tools

Joshua Leung (aligorith ATTA gmail.com)

Synopsis

In recent years, there has been a growing trend towards the adoption of more pose-centric tools for animation, inspired by workflows of traditional 2D animators. This project aims to capitalise on such trends, building upon our existing Pose Push/Relax, Propagate, and Motion Path visualisation tools, with exciting new methods for more visual manipulation and representation of animation.

Benefits to Blender

As a result of previous development efforts during 2.5, Blender’s animation system now boasts a toolset equivalent to and/or superior in some aspects to other “industry standard” stalwarts which had been stagnating for years. This has lead to (indirectly, though not necessarily without some impact) recent feature innovation from Maya most notably, with interactively editable motion paths in the 3D Viewport.

Development of these tools within Blender not only provides Blender users with modern, state-of-the-art tools for creating and editing animation in easier and more efficiently, but also acts to encourage renewed innovation in the animation tools of 3D content creation packages.

Deliverables

There are several potential technologies under investigation for inclusion into Blender, pending feasibility considerations regarding implementation matters and/or desirability by the Blender animation community. These include:

• Pose Sculpting – Just as there are sculpt tools for shaping meshes by pushing and pulling vertices with an intuitive interface paradigm, there are unexplored possibilities for allowing such an interactive brush-based painting approach to sculpting the poses of bones. In some cases, this could save time and effort required to achieve a desired shape, by reducing the amount of select+tweak+constraint+tweak cycles needed.
• Pose Sketching – Although more experimental than Pose Sculpting, time-allowing, it would be interesting to implement some of the current methods in this relatively new field of investigation to evaluate their effectiveness with actual users and production rigs.
• Motion Path Editing in the Viewport – Editing the shape of arcs and/or timing within the viewport would greatly simplify the workflow for animators when polishing their animation, providing a more intuitive interface than via a curve editor.
• Spacing Charts – While potentially less useful with Motion Path support, these 2D animator staples may well be very useful for transitioning users and/or for animation students mastering the art of timing. Besides, they are very nice eye candy that is probably more useful than the “view cube” turns out to be!
• “Synaptic” View – A rigger-defined graphic display used to facilitate easier selection and/or management of animation controls. This has been anecdotally claimed to provide significant workflow improvements, in terms of reducing time wastage on scanning a chaotic + frequently changing visual search space for controls to select.

Project Schedule

Work can begin immediately although due to term-time commitments (being in a Southern Hemisphere University, and working on a year-long group project this year) there will be times when development will be carried out in short sprints (usually over weekends). There will also be an exam period during late June (dates are still unknown due to post-quake chaos).

Depending on the targets chosen, I’m planning on aiming to tackle two large targets, one each side of the mid-term evaluation, with additional smaller targets slotted in as time allows. I’m more keen to tackle the Pose Sculpting and Spacing Charts targets before Motion Path editing though, given that there are a few technical issues I’ve still got to think through first with the latter (namely, related to rotation+scaling handling).

Bio

I’m a third year Computer Science student at the University of Canterbury, doing my (hopefully) 3rd GSoC with Blender. As the main coder+designer behind the current animation system, I’m well aware of the issues involved in this domain. As a hobbyist animator, I’m also well aware of some of the issues which animators face, having created some unique tools for Blender such as “Long Keyframes” and “Pose Propagate”, and also the implementation of the Grease Pencil tools.