Major Project 2: Final Compilation and Reflection
5.2.2026 - 29.3.2026 / Week 1 - Week 8
Chan Wan Qing / 0350928 / BA of
Design (HONS) in Creative Media
Major Project II
Final Compilation
and Reflection
INDEX:
1. Project
2. Feedbacks
3. Reflection
MAJOR PROJECT II
This module is a continuous project from Major Project I. We are expected to come up with a final animation in the end of the semester. Our group decided to continue together with the idea from Major Project I.
Modeling, Texturing
Since we had already have the idea for our project, and we decided to create the animation in 3D. We started to model all the characters, environment and props we needed. In the first consultation section, we discussed our project timeline with Mr Kamal. We have to try to finish all the modeling in the first week, then followed by rigging in the second week. Therefore, we delegated the tasks within the members. We have 4 characters and 4 main environments and a few props. My task is to model one character, Xu Chen and an environment, the graveyard.
I used AI to generate the turnaround sheet in 3D to have a better reference to model the character in high poly in Blender. I used the character model from ChaoRan as the base and model the body following the reference.
After modeling the body to match with the correct proportion of Xu Chen, I modeled the shirt and trousers with 2 meshes respectively. I used knife tool to create the pocket on the shirt and extrude tool for the belt. In this process, I mostly used subdivision, extrude and scale tool. For the hair, I added a few meshes and model them following the reference.

Xu Chen - Face Structure (Refine)
Then, I also modeled the lantern and added light in the lantern.
Rigging
1st Attempt
After modeling, we started the rigging process. As what we researched in Major Project 1, I used Rigify to rig the character. At first, I added the human bone from Blender. Then, I adjusted the bones to match with the character and generate rig.
2nd Attempt
I faced some issues where I cannot generate the rig. And that is because I deleted a few bones in the facial area. Therefore, I deleted the armature and redo the whole process. At last, I successfully generated the rig.
However, when testing the bones, there are a few issues. For example, the spine is not following the body; the eyelids are not following the rig. Also, we can't find the neck bone to do weight painting; spine bone shows weight but doesn’t have any influence on the meshes
Therefore, I tried another method, which is using AutoRig-Pro. After matching the bones with the meshes, I generated the rig and bind the rig with the meshes.
Initially, the face area's rig is not working fine, but the issue is fix by rebinding the rig with head mesh in heat maps. Another problem is that the knee is bending backwards, and this is caused by the direction of the knee joints. I have to bend the knee joint to the front a little bit before generating the rig.
Compiling Environment and Characters
After finishing all the modeling and rigging, our group started to compile all the elements together. We tried importing the characters to the environment with obj., but the character's rig is not following. Therefore, we tried to append the characters into the environment. After appending the characters, we have to match the size of the environment with the characters. Then, we uploaded all the compiled scenes into a drive to make work easier for us when we started to animate.
Animating, Lighting
After that, we delegated the scenes we have to animate. There are 4 pages of the storyboard according to our Major Project 1's plan, so each of us took a page. I started animating according to the storyboard. First, I adjusted the angle of the camera. Then, I adjusted the position of the character and started animating. For the ghost, I mostly adjusted the head and spine controls for character movement. I also applied stretch and squash principles to enhance motion of the ghost.
For walk animation, I imported a walking reference and pose the characters. To make the walk animation looks natural and smooth, I also adjusted the animation using graph editor. I have to also imagine the possible movement of the characters when they are speaking according to the designed dialogues.
While animating, I had to add some effects. Therefore, I found some online tutorial for help in adding effects. For example, I added glow effects using emission. I also learned to animate inflating using shape keys and sculpt tools. I also added sky in the environment. Besides, I also learned to append action to the character if the character is doing a same action such as walking.
Since the environment file for my scenes already have lighting, I only have to add some filled light for the characters since some of them look too dark. I tried to make the light softer so that it will not create harsh shadow on the face of the characters.
Rendering
After finishing animating, I started rendering the scenes. We tried both eevee and cycles. Although cycles took longer time to render, we all agreed that cycles look better. And in eevee, we can't render the sky. Therefore, we rendered all the scenes in cycles. I took 3 days to render all the scenes.
Editing
After we rendered all the scenes, we uploaded to a drive. ChaoRan and I will do the editing. We delegated the work among ourselves. Each of us will edit 2 pages of the storyboard. I cut all the extra scenes and adjust the pacing for each scenes. Since we have to keep it under 4 minutes, we tried to cut all the unnecessary scenes and speed up some scenes. Then, I did the end credits according to the Gann Chart. After that, I added the sound effects and voiceover. I found some royalty free music online but they don't really match with some parts of the final animation.
After presenting, I changed the voiceover with text-to-speech function in CapCut. After exporting the voice over, I edited them in Premiere Pro. I also adjusted the pacing of the final part because it is too fast. After each of us is done with the final amendments, I re-edited the final animation.
Final Submission
1. Project Exploration
Project Management Document: https://docs.google.com/document/d/1p18kmq1cIEiKQSyAncFvqd4KBxbUWWGKlthBwzw2eOE/edit?tab=t.3qdfhmnr05qi
2. Project Execution
Google Drive Link:
3. Presentation Slides
4. Final Animation and Art Bible
FEEDBACKS
5.2.2026 / Week 1
- Plan properly before starting the modelling and animation process to ensure
a clear workflow and efficient task distribution.
- List all required
characters, props and environments needed for the project to avoid missing
assets during production.
- Organise the production timeline using a
Gantt chart to manage deadlines, monitor progress and ensure the project stays
on schedule.
- Prepare a detailed shot list based on the storyboard to
guide the animation process and maintain consistency in visual
storytelling.
- Assigned specific modelling roles among team members to
ensure balanced workload and efficient production flow.
- ChaoRan
presented her initial character modelling work, Mr Kamal advised using it as a
base body model as a foundation for other characters so that we can do work
more efficiently.
- Complete the entire modelling phase within one week
to allow sufficient time for subsequent stages, including texturing, rigging,
animation, lighting, rendering, compositing and editing, ensuring smooth
project progression.
12.2.2026 / Week 2
- Eyelids need to have upper and lower
- Add a few more subdivision on
joints (fingers, elbow, knee)
- Lips shape
- Eyeball has to be
round
- Head shape needs to be smooth
- Keep the fingers straight
(rigging easier)
19.2.2026 / Week 3
- Add UI bone collection to generate rig.
- For the facial, it’s okay to
use the current facial, just a lot of weight paint to do especially on the
eyes.
26.2.2026 / Week 4
- Overall is ok
- Just the eyelids is not following the eye area
-
The spine bones are not working
- Problem: Couldn’t find the neck bone
to do weight painting; spine bone shows weight but doesn’t have any influence
on the meshes
- Export the animation as a video file for consultation review to ensure
accurate frame rate playback, as the Blender file is too heavy to run smoothly
in real time
- For repetitive actions such as walking cycles, use the
append animation function to reuse existing animations and improve workflow
efficiency.
- Set up the camera before starting the animation process to
ensure that all animated elements remain within the frame and align with the
intended shot composition. Avoids animating elements that will not appear
within the camera view to save more time.
- Ensure that character actions
and movements correspond with the dialogue, allowing the animation to
better support the narrative
- Adjust the arm movements and finger positions to appear more natural, as
they currently look too stiff
- Increase the speed of the ghost turning
scene to improve the pacing of the action
- Avoid keeping characters
static for too long; add subtle movements to make the scene feel more natural
and lively
- Ensure the camera remains on the same side when showing two
characters in conversation to maintain continuity and avoid confusing the
audience
- When a character turns their head, ensure that the body
movement follows naturally to maintain realistic motion
- Adjust the
pendant animation so that it opens more slowly, creating a smoother and more
controlled motion
- Reduce the speed of grass moving as they move too
fast that they are more eye catching than the characters
- Sound Effects, VO, BGM (Front part too quiet)
- Add title in front,
differentiate final scenes and ending credits (Ending credits animation change
to b&w)
- Make it slower (last part)
- 1st scene, metal door
draw too much attention
- Sky colour (first part) - brighter
-
Hell scene - door, light at both side
- Fire effects
-
Transformation part - light
REFLECTION
This has been the most intense 7 weeks throughout my three years in university. Although we had already planned everything during MP1, the execution phase in MP2 pushed me to my limits. From modelling characters, environments, and props, to rigging, texturing, animating, adding visual effects in Blender, and finally editing, every step was equally important and demanding. I was able to apply the knowledge I learned over the past few years, while also picking up many new techniques that I had never explored before.
One of the key challenges was maintaining a consistent visual style across all assets within such a short timeframe. Due to time constraints, we were not always able to refine and match every element perfectly. Technically, I faced difficulties in Blender, especially during rigging (incorrect mesh parenting, hair not following the head, problems with Rigify which I eventually used ARP) and animation. I also learned new techniques such as using shape keys for deformation effects and animating emission for glow effects. One aspect that surprised me was how much thought goes into body acting and character performance. Designing believable gestures and expressions for the characters while they speak was more harder than I initially expected, and it played a significant role in enhancing the storytelling. During the editing stage, another challenge was keeping the animation within the 4-minute limit while maintaining a natural narrative flow. Cutting scenes required careful consideration to ensure that the pacing felt smooth and that the dialogue remained well-synchronized with the visuals.
Despite the challenges, completing the final animation within 7 weeks was very rewarding, and I am proud of what our group achieved. Moving forward, I would like to further improve the quality of my animations, particularly in character acting and expression, to make them more natural and engaging, and also improve my workflow.


.png)





































Comments
Post a Comment