Introduction
Computer-assisted cel animation (CACAni) has been carried out since 1996. The objective of this project is to design and apply computer graphics and imaging techniques to develop a generic 2D animation system which can be used in entertainment industry and multimedia production. Specifically, the aim of the project is to develop cutting-edge technique that aims at optimizing the image-related working process by introducing computer-support in the drawing and painting stages of traditional cartoon production, in particular, allowing automatic in-between generation, painting and colouring of a sequence of digitized/vectorized images.

Computer-Assisted Coloring by Matching Line Drawings (Seah Hock Soon and Tian Feng)
An approach to automatically color line drawings based on feature matching is proposed. The motivation is that coloring 2D animation is still a labor-intensive process in current cartoon film production. The objective of our work is to investigate how to automatically color an image in a cartoon sequence based on its previous frame. Our method first establishes matching relationship of two images, which yields a set of displacement vectors. After which it automatically paints one of them based on the color information of the other using a region matching algorithm. The region matching algorithm is based on the set of displacement vectors obtained earlier. The results show that the proposed algorithm can straightforwardly and robustly realize our objective and has promising future for our next step to further automate conventional animation.

Examples:
1. Bear sequence
2. Stomach sequence
3. Angel sequence

Computer-Assisted In-between Generation (Seah Hock Soon and Lu Ji)
An approach to automatically generate in-between line drawings based on feature matching is proposed. The motivation is that producing in-between frames in 2D animation is a labor-intensive process in current cartoon production. The objective of our work is to automatically generate the in-between frames between two given key-frames. Our method first determines matching relationship of the two images based on feature correspondences. After which, it automatically generates in-betweens using an interpolation algorithm. 

Examples:
The following is a test sequence we work with Jaymen from the LaSalle-SIA College of the Arts:-
1. Original key frame sequence.
2. Interpolated sequence with 3 in-between frames inserted between every pair of key frames.

The following is a test sequence we work with Rafael from TouchToons Animation Studio:-
Generating inbetweens

Autocoloring Based on Master Frames
The scope of this project involves auto-coloring of character based on master frames. Student will be provided with existing CACani image correspondence and auto-coloring modules. Together, these modules allow users to input hand-drawn cartoon frames (cels) to automatically color the sequence based on the manually colored first frame. This is the basis for student to carry out the investigation. Initially, the student is expected to be familiarized with the capability of the existing auto-coloring algorithms and the programming environment. Using region matching, the algorithm attempts to color the subsequent frames of the input sequence based on the color information of the first frame. This technique has the problem of missing information due to occlusion, that is, the appearance and disappearance of regions between frames. In this project, we are researching a computer method to mimic the way human applies colors to individual cel based on a set of master frames which define the appearance of individual character to be animated. With these master frames, human will be able to color the character in any of the postures it appears. Will the computer be able to do the same? Student will investigate this problem using the existing auto-color algorithm and improve on it if necessary.

Animation of Natural Phenomena
The scope of this project involves the animation of natural phenomena like water, cloud, fire, etc. which are difficult to animate using traditional hand drawings. Student will be provided with existing CACani image correspondence and in-between generation modules. Together, these modules allow users to input key frame drawings to automatically generate an animation sequence. This is the basis for student to carry out the investigation. Initially, a study on the current methods of animating natural phenomena with hand drawings is expected while student is familiarizing with the capability of the algorithms and the programming environment. Based on the results of the information gathered, student will attempt to automate the various phenomena using the existing CACAni system. Results of the automatic animation sequence are to be observed and analyzed. Based on the analysis, student is to recommend improvement to (a) the ways input key frames are to be drawn and (b) the existing algorithm to handle animation of natural phenomena. If time permits, program suggested modification to the existing algorithm.

Cel Shader
This portion of research looks into rendering of cel uisng techniques like mapping (texture, pattern, bump), color shading, depth cue and variable line weight and reflection.

Lip Sync
Application of digital image processing on recorded video to generate lip movements.

Useful Link: Illustrate! A market-leading cel and illustration renderer.