Sketch and Animation Assistants for Novices

PhD Thesis Proposal Defence

Title: "Sketch and Animation Assistants for Novices"


Mr. Qingkun SU


Sketch and animation are two frequently adopted art forms for expressing 
ideas and conceptions. However, even for professional artists, producing 
visually pleasing results requires substantial expertise and effort. 
Because sketch and animation are difficult to master, drawing assistant 
tools are necessary for novices. However, existing commercially available 
software packages for sketch and animation, such as Adobe Photoshop, Toon 
Boom, demand a steep learning curve and are rather difficult to master for 
novices, further hampering users' creation process. In this thesis, we 
present two user-friendly assistant tools, EZ-Sketching and Live Sketch, 
to ease the creation process of sketch and animation respectively for 
novice users. We also propose one error-tolerant target acquisition 
technique, 2D-Dragger, to solve the common accessibility problems in 
existing tools on touch devices.

EZ-Sketching, a novel image-guided drawing interface, uses a tracing 
paradigm and automatically corrects sketch lines roughly traced over an 
image by analyzing and utilizing the image features being traced. While 
previous edge snapping methods aim at optimizing individual strokes, we 
show that a co-analysis of multiple roughly placed nearby strokes better 
captures the user's intention. We formulate automatic sketch improvement 
as a three-level optimization problem and present an efficient solution to 
it. EZ-Sketching can tolerate user errors from various sources such as 
indirect control or inherently inaccurate input, and works well for 
freehand sketching on touch devices with small screens.

Similar to the image-driven assistance in Live Sketch, our animation 
assistant tool, Live Sketch, allows novice users to interactively bring 
static drawings to life by applying deformation-based animation effects 
that are extracted from video examples. Dynamic deformation is first 
extracted as a sparse set of moving control points from videos and then 
transferred to static drawings. Our system addresses several major 
technical challenges, such as motion extraction from video, 
video-to-sketch alignment, and many-to-one motion-driven sketch animation. 
While each of the sub-problems could be difficult to solve fully 
automatically, we present real-time and reliable solutions by combining 
new computational algorithms with intuitive user interactions.

In addition to improving users' drawing experience and aesthetics, we also 
present a target acquisition technique on touch devices, 2D-Dragger, which 
aims at making selection of UI elements easier and more accurate for 
users. Occlusion (targets are small and dense) and inaccessibility 
(targets are out-of-reach) are two main problems that may lead to 
inaccuracy and inefficiency. Our main idea is that the effective width of 
each object is constant and same, allowing a fixed scale of finger 
movement for capturing a new target. Our tool is thus insensitive to the 
distribution and size of the selectable targets, and consistently and 
robustly works well for selecting targets with various distribution.

With the above three assistant tools, we expect novice users can quickly 
and painlessly create drawings and animations. To evaluate the 
effectiveness and robustness of the three techniques, we conduct several 
user studies for each technique. The results show that our methods allow 
users without drawing or animation skills to easily create plausible 

Date:			Thursday, 14 December 2017

Time:                  	3:00pm - 5:00pm

Venue:                  Room 5501
                         (lifts 25/26)

Committee Members:	Prof. Chiew-Lan Tai (Supervisor)
  			Prof. Huamin Qu (Chairperson)
  			Dr. Xiaojuan Ma
 			Prof. Long Quan

**** ALL are Welcome ****