During the third period of my 3D animation study at HKU I started working with a humanoid rig. The rig I used is called Bony and is free to use for everyone who wants it. I like rigging but we weren't allowed to rig a character ourselves for this assignment. During this course we got to rotoscope four different kinds of movements or types of footage. I really liked the assignment and I felt I had more to learn from it so I decided to do some of the assignments twice or longer than what was really necessary.
For the first exercise we were to rotoscope a 12 frame walkcycle with the images of Eadweard Muybridge as reference material. I was curious to see if there we a very noticeable difference between a man en woman walking, that's why I decided to rotoscope both. For the second exercise we had to rotoscope a jump, still making use of the Muybridge images. At first glance I thought that the first jump I wanted to rotoscope would be really easy and so I wanted to do a second more difficult one. In the end the first jump was actually much more difficult than the second one. This was, in hindsight, because the movements in the second one were much more exaggerated and more easily visible than the first one. This was a very interesting lesson to experience. For the third exercise we got to rotoscope live action footage. It was actually quite a challenge to find live action footage without a moving camera, until then I never really took notice of how much a camera moves around. In the end I decided upon a small part of a fight sequence. Here I ran into Bony's rig limits. It was very difficult to mimic Bruce Lees movements using Bony but I think I did alright. We only had to rotoscope 12 frames but I really wanted to finish the swipe so I rotoscoped the whole movement (which was 24 frames). And for the last exercise we got to choose what type of footage we wanted to use, we could choose between 2D/3D animation, an insect or an animal. I always really liked the cartoon Avatar, Toph was/is my favourite character, so I decided that this would be my final rotoscope. I think I got the movement alright but I had a lot of trouble aligning the camera properly.
During this whole assignment I got a better grasp of how the human body works, with it's counterbalancing between the hips and shoulders, how the human body is never (never!) still, how it's never really symmetrical and how to shift weight and balance. I've said it before but I really enjoyed this assignment and I learned a lot from it. The music in the video is: Ghost Fight (UNDERTALE) Jazz Cover.
This project is the final project for the second period of my 3D animation study at HKU. This was a real treat to make! We had complete freedom regarding what we wanted to do so long as it was Christmas themed and the video shouldn't be too long. I'm not much of a Christmas person myself but that's actually why I wanted to give these cube guys a nice Christmas experience.
First I started with writing down some ideas and the story came together pretty quickly. I wanted it to be a propper three act story structure with an "all's well that ends well" ending, a little cheesy maybe but it suits the Christmas theme in my opinion. Once I finished writing the story I stared on scribbling down the storyboard. Before this project I had very little experience with digital drawing it was an extra bonus I got to put some time in on Krita (the program I use for digital drawing). I recorded my own voice as voice-over for the characters, this was pretty scary for me but in the end I had a lot of fun doing it, and I collected all the necessary sound effects off the internet. I put everything together as an animatic using Premiere Pro.
Before I started animating I asked my classmates and my teacher for some feedback to improve on the storytelling. Once I implemented the feedback I stared on creating the world and modelling al the objects. Since the main characters were just plain old cubes I decided to keep the environment as simple as possible regarding both shape and colour. I've always enjoyed a low poly look so I was happy to be able to use it once more in my latest animation.
After I finished creating the environment I saved the file with three separate names for the three separate scenes I had decided on. I also created a fourth file where I used a particle system to populate the mountain slope with pine trees. It was easier for me to animate the scene where the parent cube comes rushing down the mountain on top of the tree that way. With all the scenes in separate files I added in the lights and camera's and I blocked out the general animation. After some testing and re-blocking I started animating.
I've come to really enjoy character animation over these past few months and I was really glad I could put all the lessons I had learned in the previous exercises to good use in this animation.
When all the animations were done I rendered the shots , re-recorded the voice-over where needed and compiled the along with the music and sound effects using Premiere Pro. While I was working on the audio I learned a lot about the timing of a sound. I created this animation using Blender 3.00 and it's rendered using EEVEE.
Concepting:
- Writing the Script
- Drawing the Storyboard
- Research on music and sound effects
- Record voice-over for both characters
- Editing the Animatic
- Implementing feedback throughout the process
Modelling:
- Modelling the house and all the furniture
- Modelling the axe, saw and chainsaw
- Creating the whole scene and adding in trees from previous projects
- Created a separate scene for the sliding sequence where I used a particle system to place all the trees
- Rig the parent and kid cube characters
Creating the Set:
- Adding camera's on locations to create a flow for the animation
- Add lights
- Blockout the whole animation to get a good feel for the timing
Testing:
- Light intensity, placement & colors
- Camera flow
- The timing and blockout animation for each scene
Animation:
- Animate the parent character in scene 1 (storyboard image 1 - 10)
- Animate the kid character in scene 1 (storyboard image 1-10)
- Animate the parent character in scene 2 (storyboard image 11-27)
- Animate the kid character in scene 3 (storyboard image 28-42)
- Animate the parent character in scene 3 (storyboard image 28-42)
- Rendering
Editing:
- Video editing in Premiere Pro
- Re-record voice-over for both characters where needed
- Audio editing in Premiere Pro
Here you can see the house I modelled and (most) of the furniture models inside the house. I also added the snowman outside the house to underline the winter/Christmas feeling.
When the parent tries to cut down the large pine tree on top of the hill most of his tools break. I decided to make separate models for these broken tools instead of using BlendShapes, this is a better fit for the style I was going for.
As I mentioned above I used a particle system to create the environment for the sliding sequence. Above you can see all the trees that were needed to properly fill the scene. You can also see that the whole plane is tilted. It was easier to animate the plane moving from down to up that have the cube and tree move downwards.
Since we were only allowed to use cubes as characters for this animation there weren't a whole lot of rigging options. So I kept it simple; controllers for the "feet", controllers for the "hands" and a controller for the "head". I adjusted the head controller a little so it will stay in the centre of the feet when moving the feet, of course the head can be moved on it's own using the controller.
During the second period of my 3D animation study at HKU I finally got to work more in actual 3D animation. Our first assignment was to create a convincing walkcycle using a cube. While the exercise seems pretty simple I found it actually holds a lot of valuable lessons. While making these walkcycle animations I started to get a better feel for some of the animation principles like Timing, Squash and Stretch and Exaggeration.
I also learned a lot about the weight of a character and how/where to displace it to when moving. I learned this partially by observing reference materials but mostly by acting the action out myself. Which is also something I never used to do but I found that it's a very helpful tool to make an animation more lifelike.
There was a lot of freedom regarding the type of walkcycle we were required to create. That's why I decided to make multiple different ones instead of just focusing on just one regular walkcycle, making sure to challenge myself with each new animation. I had a lot of fun acting out these animations and of course also while creating them!
This regular walkcycle is the first animation I made for the assignment. I also tried to slow it down to 130 fps and speed it up to 65 fps, just to see what it would look like. But I like this version best.
For this next walkcycle I tried to challenge myself a bit more by displacing the weight more. I also tried to express some emotion (pain and effort) during this animation.
Here you can see that the characters walk is being influenced by forces from the outside, snow in this case. This makes his steps slower and he has to put more effort in to move forward.
For this exercise we had to add two walkcycles together by an event somewhere in the middle. I used my snow walk and my regular walk. For this animation I also focused on the event in the middle, trying to make it a little amusing. I also got a better grasp of the Squash and Stretch principles during this exercise and took special care to have the feet jump independently from each other, this is more natural.
This walk was both the most fun and also the most difficult one I had done up till this point. I used reference material I had found on YouTube but I mostly acted the animation out myself. I was a challenge to get the timing right and have the weight shift back and forth in a convincing but not too stylized manner. I tried to push the poses where possible and I tried to make it as entertaining as I could.
During the second period of my 3D animation study at HKU we got the assignment to make a short lipsync animation. At first I tried another version of motion capture (like I did with the Poem animation a bit further down this page) but I wanted to challenge myself a bit more and decided to do another lipsync animation. This time I decided to use BlendShapes.
Since I had only used facial BlendShapes once before, in one of the Eel animations I made for Virual Play (also a bit further down this page), it was a pretty new and really fun experience. I learned a lot about the shapes our faces and mouths make when speaking, I hadn't given it much thought before this assignment.
Later, while working on the animation, I discovered that the style I had used to make the Blendshapes (big en wide gestures) didn't really suit the type of audio I was using. So I adjusted the shapes a little to make them a little more realistic, this was a lesson I did not expect to learn but a very valuable one nonetheless. It's definitely something I will be paying closer attention to in the future.
For this assignment I decided to focus my attention on the animation rather that the modelling. So I used models I could find online to create a character that slightly resembled the character in the scene from the audio fragment.
Once I finished the lipsync I rendered the shot and compiled the along with the audio, the audio you're hearing is a fragment from WestWorld season 1 episode 10. I created this animation using Blender 3.0 and it's rendered using EEVEE.
Concepting:
- Researching Blendshapes
- Researching 3D mouth shapes and phoneme mouth charts
- Acquiring audio fragment
Modelling:
- Get propper models from the internet and add teeth, eyes and eyelids to the head model
- Create each individual BlendShape
- Add controller bones and drives so each BlendShape can be animated
Animation:
- Listen carefully to the audio and write down the phonetic words
- Select which emphasis should be animated an which should be left out
- Animate the words/emphases using the BlendShapes
Creating the set:
- Use the models found online to create the whole set (slab, robot body and hair)
- Add and animate a light to support the audio
For one of the first animation assignment for my current study, 3D animation at HKU, I made the animation you can see above. It was a class assignment and the full result can be found here. We choose a song, 19-2000 by Gorillaz (Soulchild Remix), and we dividend the parts amongst the class. The theme was "vacation" and since I used to love skateboarding during the summer holidays this seemed like a really fun idea to use in this animation assignment.
It was a really fun assignment and I really loved working on it. I learned to work with new shaders to give it that 2D neon look and I also dove a little further into the lighting settings for the 3D animation shot.
I had a lot of freedom regarding the content and style of the animation. I decided to go with a low poly style, to give it that "easy going" feeling that the character should have during his vacation. I started with writing a short script for the animation and after that I drew the storyboard, which you can find that below, and made an animatic. I also started on the modelling, texturing, rigging and weight painting of the character (I named him Guy) and the skateboard.
Once the character was done I started on his poses for the animation and later on I started working on the actual animation. And I created BlendShapes for the shirts so it won't clip through the legs and it'll move up when the character jumps. I also created the rest of the models, the trees, texts, grass, stones etc.
The first and third shots I animated in 3D first, later I added a shader which shades the whole model black, adds a solidify modifier and shades these extra faces in a selected color. But since there are no lights in these shots everything is flat and looks 2D. For the character I also animated the colors, to make the scene look move vibrant and in action.
Once all the animations were done I rendered the shots and compiled the along with the music. I created this animation using Blender 2.93 and it's rendered using EEVEE.
Concepting:
- Research real life skateboarding
- Write the script
- Draw sketches of the main poses in the animations
Modelling:
- Model the character and skateboard
- Model the environment assets
- Texture all assets
- Rig the character and skateboard
- Check and/or adjust the weight paint
Creating the Set:
- Adding camera's on locations to create a flow for the animation
- Add lights
Animation:
- Rig the model into the poses I wanted to use in each animation
- Animate each individual animation using the sketches and poses
- Shirt animation: create BlendShapes so the shirt moves along with the legs
- Rendering & compositing
As you can see, Guy is a pretty low poly looking character. But instead of sharp edges I decided to go with a more smooth and rounded look for the character.
Here you can see the character rig, it's a pretty basic rig but it worked really well for the animation.
For the skateboard model I wanted to stay as close to a real skateboard as possible while also making sure it still fit the style. The deck and trucks are separate parts so there wouldn't be any troubles while animating.
While rigging the skateboard I had to make sure the I could move the wheels and the board separately while also being able to move and rotate the whole thing at once. The rig I came up with works but I'd definitely like to improve on this rig in the future.
Guy - Side View
Guy - 3/4 View
Guy - Close Up
Guy - Extreme Close Up
Environment models, all background models in one image. Clouds, broadleaf and pine trees, rocks, bushes and grass.
Environment models, different colored broadleaf trees. I wanted to change things up with the colors, not just green.
Environment models, pine trees. Also with these trees I wanted to change up the colours, though these colours are a little darker than the other trees.
Environment models, rocks. I used these rocks to hide the edge between the ground and sky and to fill up the background.
Environment models, stairs and road. Since you can't skate very well on grass so there would have to be some sort of road for him to ride on. Guy also make a big jump over some texts so he would need to have a platform to jump off from. That's why I made these supporting models.
The skill animation is important for the game so the eel Hatchling can help the player during the pinball game. Because it's used in the game the animation needs to be pretty quick and clear on what the eel is doing. For this animation I wanted the eel to be a little funny so I made him jump off his pedestal like a coil. But I also wanted to remain close to the real thing so the animation is fluid with nice round forms and movements.
For this animation I made 2 different BlendShapes. One for the startled look at the start of the animation and the other is a very exaggerated emotion to show how very sad he is. He'll start moving more slowly and his whole body starts to hang. This to really push the emotion. He does however need to return back to normal at the end of the animation so he can't stay sad. The eyes will be animated using code in the game.
Eels are pretty aggressive animals, especially when they're eating. But for this eel it's very important for him to stay cute while also be realistic. He does attack his food, a little, and shakes his head when eating but not as aggressively as an eel does in reality. For this animation I built anticipation before he attacks his food to make the animation realistic and interesting.
This animation might actually be as close to realism as I got when making these animation. After studying videos of divers and moray eels it turns out that eels(the ones that are used to human at least), much like dogs, lay their heads in a handpalm when one is presented so it can be pet. It is very important for this animation that the eel looks cute so I rotated his head upwards pretty far and made him hand his head from side to side, imitating a dog.
Above you'll find some of the animations I made during my internship at Virtual Play. These will be used in the game they are currently developing; Hatchball.
Hatchball is a combination of the classical game Pinball and 3D Tamagotchi. During the game the player can collect eggs for Hatchlings (the animals in Hatchball) which the player can hatch later on in a hatchery. Once hatched the animals will help the player during the pinball part of the game.
For this game a great deal of animals (Hatchlings) had to be created and animated and so I did most of the animations for the Electric Eel.
For the Eel it was very important that the animations stayed as close to the real thing as possible while it should also be cute and cuddly. Of course this wasn't an easy task since a real eel isn't particularly cute or cuddly and of course it also doesn't show any actual facial expressions. So for Sad Animation facial expressions I used Pixar movies as inspiration and pushed the expression further. While the Love and Eat animations are kept closer to reality, for these I studied videos of electric eels and moray eels.
Since the eel is a very smooth and flexible animal I took extra care so that all its movements are also be very smooth, round, flexible and effortless.
The eyes and eyelids aren't animated in most of the videos above because these will be animated through code in the game.
I made these animations using Blender 2.9 and they're rendered using EEVEE.
Concepting:
- Research real life video footage of eels
- Draw sketches of the main poses in the animations
or
- Rig the model into the poses I wanted to use in each animation
- Document the sketches/images and describe each action in the animation through text
- Consult with Art Lead and make sure the ideas are green lit before starting
Modelling:
- Make sure the rig works well
- Check weight paint for any odd or misplaced weights
Creating the Set (if it's needed to better visualise the animation):
- Add food models for Eat animation
- Add "pedestal", water and a ball for the Skill animation
- Adding camera's on locations to create a flow for the animation
- Add lights
Animation:
- Animate each individual animation using the sketches or poses
- Sad animation: create BlendShapes for the facial expressions
- Rendering
Feedback:
- Consult with Art Lead if animation needs to be improved
Unity:
- Import the model and animations into the Unity project so it can be used in the game
Electric eel - side view
Electric eel - 3/4 view
The model and texture were made by a number of other interns, under the guidance of the art director of Virtual Play.
During my internship at Virtual Play I was asked by a different company, Remark, to create assets for his new game DigiShi. This was of course an awesome opportunity and it is also how I decided to create my own studio, Fancy Fox Studios.
The first step of course was to have a meeting and discuss what the customer was actually looking for. DigiShi is a VR game for the Oculus Quest where the player can program his sheep, using cards, to get to the other side of the gameboard while also learning more about sheep by answering questions. At first the customer wasn't sure about the style he was looking for. He did know he wanted the sheep te be programmable, which means that they should have some mechanical detailing. I provided him some moodboards and in the end we settled on a low poly style, this is nice since the poly count won't be too high so the Oculus won't have any trouble running the game and it's a very child friendly style (young players are his main target audience).
Once we settled on a style I started drawing the characters and later modeled them along with all the environmental assets. I had a lot of freedom regarding the animations. The customer knew which animations he wanted but he wasn't sure on how they should actually look. So I drew storyboards for the animations, which the customer liked very much. For the animations I wanted to make sure they were comical and theatrical. Like the Wolf bending extra low when sneaking forward or the Sheep jumping sideways in a very theatrical manner. I decided on these exaggerated animations because of the style we landed on but mostly because they should be fun to look at for the targeted audience.
Once all the models were rigged and weight painted I started animating them using the storyboard. All these individual animation can be found below.
I made these animations using Blender 2.9 and they're rendered using EEVEE.
Concepting:
- Customer consultations and meetings to figure out what he is looking for
- Create moodboards and consult with the costumer again, see how he likes it
- Create character concepts
- Draw storyboards for all the needed animations
- Documentation
- Customer consultations before I start modelling and animating
Modelling:
- Model the characters (Sheep, Wolf and Mole)
- Model the environment assets (tiles, trees, bridge etc.)
- Create a colour atlas
- Consult customer
- Rig the characters
- Weight paint the characters
Testing:
- Test the rig
- Test and check the weights
Animation:
- Animate Mole animations
- Consult customer
- Animate Sheep animations
- Consult customer
- Animate Wolf animations
- Consult customer
- Animate the promovideo seed above (for portfolio purposes only)
- Rendering
Feedback:
- Consult the customer and see how he likes the models and animations and change anything that isn't up to snuff.
Unity:
- Export the models and animations to a Unity project and export this new scene as a .unityproject file. This so the Game Developer can import the file and use the models and animations
Editing:
- Video editing
Jump forward/backwards
Jump left
Jump right
Turn left
Turn right
Idle, shake leg
Idle, look left right
Idle, wobble left right
Idle, sniffing
Shake head
Success
Fail
Idle, howl
Sneak forward
Idle, sniffing
Turn left/right
Come up from hole
Go back down hole
Sheep - front view
Sheep - side view
Sheep - 3/4 view
Wolf - front view
Wolf - side view
Wolf - 3/4 view
Mole - front view
Mole - side view
Mole - 3/4 view
As my final project for the extra Animation course I took during my current studies, I was required to make a 30 second animation. I set out to look for a company that could use such an animation and I found Aryzon. Aryzon wanted an introduction animation to their new product, Aryzon World. They asked me if I could help them come up with a concept to show how useful their product could be in the educational sector. So I did some research and found out that most teacher have a hard time teaching because there is no interaction with their students. With this in mind I wrote up a concept and made some sketches. After a brainstorm session with the Aryzon team a few adjustments were made to the initial concept and so the final script and storyboard were created.
For this animation I decided to go with a low poly look. It is what appealed most to the customer and it is what was best suited for the timeframe I had to finish this exam project. Since the modelling would be low poly, the colors and materials would also have to be basic to give the whole scene the same look and feel.
For the first scene I created a dark, glum and empty looking classroom to better represent how it sad and demotivating it is for him to teach with no one present. The second scene is the complete opposite. The classroom is bright, with happy colors and the schooldesks are placed in rows instead of 1,5m apart from each other. The classroom is also full of animated 3D headset representing students. The third scene was somewhere in the middle of the first two. Since the teacher was still in the empty classroom, it shouldn't look too bright and joyous but the teacher is happy to use the product and he is happy to teach digitally. So the lighting and colors are a bit brighter.
I made this animation using Blender 2.9 and it's rendered using EEVEE.
INT. HAL EN KLASLOKAAL
Hal en klaslokaal zijn leeg, geen leerlingen te vinden. Alle tafels en stoelen staan op anderhalve meter afstand van elkaar. Weinig kleur, lage lichtintensiteit.
1. Leraar loopt vanuit de hal naar een leeg klaslokaal
- Voetstappen –
CAMERA kijkt over schouder mee
2. Leraar staat een moment stil in de deuropening en
kijkt naar de lege tafels
CAMERA kijkt over schouder mee
3. Leraar loopt naar binnen en draait met
hoofd naar lege tafels
CAMERA over schouder van leraar mee
totdat leraar een stukje het lokaal in is
4. CAMERA gericht op leeg klaslokaal en
draait langzaam rond z-as
CAMERA gericht op leraar
5. Leraar loopt met gebogen hoofd en zucht diep
LERAAR
*zucht*
Leraar is gedemotiveerd en bedroeft,
hij vind het maar niets om digitaal les te moeten geven.
Hij mist de interactie met zijn leerlingen
- Zucht -
6. Leraar loopt naar bureau
CAMERA kijkt over schouder mee
Leraar staat achter bureau
CAMERA is pov van leraar
Leraar opent laptop met Teams of Zoom open
7. CAMERA pov van leraar
- Keyboard toetsen –
8. Leraar pakt Oculus
LERAAR
*Ahum*
Leraar is benieuwd naar deze ervaring en probeert
de demotivatie van zich af te schudden en
met goed moed te beginnen. ??
- Schraapt keel -
9. Oculus beweegt naar camera
- Woosh geluid, vrolijke muziek begint -
10. Aryzon.World scherm
INT. DIGITAAL KLASLOKAAL
Klaslokaal zit vol met leerlingen in de vorm van VR headsets, netjes naast elkaar. Veel kleur en een hogere lichtintensiteit.
CAMERA is nu Oculus
11. Leraar ziet vol klas lokaal met leerlingen
in de vorm van VR headsets
LEERLINGEN
*Maken herrie en lachen*
De leerlingen vinden deze VR leeromgeving erg leuk
- Gelach en herrie van vrolijke leerlingen –
12. Leraar kijkt rond
LERAAR
*Lach*
Hij is erg blij om zijn leerlingen weer te zien en
hij is erg blij dat Aryzon.World dit
mogelijk heeft kunnen maken.
- Lach –
13. Leraar kijkt naar beneden en ziet handen
LEERLINGEN
*Shhh*
De leerlingen zien dat leraar aanwezig is
- Shush -
14. Leraar draait hand naar buiten (feature van Platform) en
selecteert iets om les over te geven
- Geluid om iets te selecteren -
15. 3D model verschijnt midden in de klas
16. Leerlingen gaan rondom het 3D model staan
CAMERA view vanuit de
klas in digitale omgeving
17. Leraar geeft les en leerlingen doen actief mee
INT. LEEG KLASLOKAAL
Klaslokaal is leeg, geen leerlingen te vinden. Alle tafels en stoelen staan op anderhalve meter afstand van elkaar. Redelijke aanwezigheid van kleuren, mee dan in het begin. Ook hogere lichtintensiteit.
CAMERA view vanuit de
klas in realiteit
(was 17)18. Leraar geeft les
CAMERA gericht op gezicht van leraar
(was 18)19. Leraar met glimlach op gezicht en
Oculus op les staat te geven voor leeg lokaal.
(was 19)20. CAMERA beweegt achteruit richting deurgat
21. CAMERA draait richting poster (Call to Action poster) aan linker kant
leraar zichtbaar in achtergrond
- Muziek stopt -
Concepting:
- Writing the Script
- Drawing the Storyboard
- Creating an Animatic
- Research on music and sound effects
- Customer consultations
- Documentation
Modelling:
- Create the classrooms
- Build all 3D models
- Rig the teacher character
Creating the Set:
- Adding camera's on locations to create a flow for the animation
- Add lights
Testing:
- Light intensity, placement & colors
- Camera flow
- The feel of each scene
Animation:
- Animate individual headsets and hands
- Character animation for scene 3
- Character walk cycle for scene 1
- Character animation for scene 1
- Rendering
Editing:
- Video editing
- Audio editing
Classroom from the first scene, it's darkly lit and the colors on the windows are dark to give the scene a sad feel. It's filled with empty chairs and tables .
Classroom from the second scene (digital). It's more colorful with a fun poster. The scene is lit more brightly and the widows (see Youtube video) are pink. the centre of the classroom is filled, digital, students.
Oculus Quest 1 headset and controllers. I had trouble getting the foam right on the inside of the headset so I had to sculpt most of the headset.
Microsoft Hololens 3D model. I made this model through hard surface modeling.
These are the hands I made and rigged for the animation. Using drivers the fingers will rotate inwards or move from side to side.
These are all the models I made for this project through hard surface modelling. All the models have basic materials, giving the whole scene a low poly look. The laptop got some extra emission materials to give it an actual laptop look.
Modelled teacher character with clothes. Clothes were duplicated and enlarged from the body after animation was done. Animations were also applied to the clothes.
Teacher model rigged using the riggify add-on in Blender.
As an assignment for my extra Animation course during my current studies, I was required to animate a poem and make a lipsync animation. For this assignment I choose to animate a poem from Revolting Rhymes written by Roald Dahl. I decided to go with facial motion capture because it is the best way to animate lip synchronisation and I'm very interested in facial motion capture. I was inspired by the Magic Mirror from Snow White and Shrek. I made this animation using Blender 2.83 and FaceGen Modeller, it's rendered using EEVEE.
Now every day, week in week out
The spoiled and stupid Queen would shout
"Oh Mirror Mirror on the wall
Who is the Fairest of them all?"
The Mirror answered every time
"Oh Madam, you're the Queen sublime
You are the only one to charm us
Queen, you are the cat's pyjama's."
Concepting:
- Write the script
- Draw the Storyboard
Modelling:
- Record the video
- Track the motion points
- Create the 3D models
Creating the Set:
- Adding camera's
- Adding lights
Testing:
- Light intensity, placement & colors
- Camera flow
Animation:
- Record the animation
- Rendering
Editing:
- Add the voice over
I placed facial markers all over my face using an eyeliner pencil. I placed the markers closer together where there would be more movement like around the lips and the eyes.
I made this video to compare the video I recorded of myself, how it looks in Blender and how the actual animation turns out. The final result of my facial motion capture animation. I used purple background lighting to create a dark and mysterious mood. And I used a bright light to light the face from beneath. This makes the face look mysterious and ominous.
I made this face using FaceGen Modeller and gave it a light from beneath to make is look ominous.
I like low poly modelling and I felt this style was a good fit for the look I was going for. I lit the outline from behind to give it a dark and mysterious look, of course I was inspired by the Magic Mirror from Snow White.
The overall look of the Magic Mirror is thanks to the lighting I used for this animation. While it's true that the lighting gives the overall scene a dark and mysterious look, the different style of the face also gives it a more comical look, like in Shrek.
For this project I wanted to get some more hard surface modelling experience and rigging experience with non human rigs and I wanted to do some different animation exercises. I'm a big Star Wars fan and I was inspired by the AT-ST when I modeled Sn-can. The colors are inspired by The Slave I from Star Wars. I only rigged the legs to be able to move, the top is a separate object so it can be rotated on it's own more easily. I ran into some problems involving vertex groups which I fixed with weight painting. I made these animations using Blender 2.83. They are rendered using EEVEE.
Preparation:
- Collect reference materials
- Decide on which animations I want to make
Modelling:
- Build the Sn-can model
- Add materials to give it a nice look
- Rig the model
Creating the Set:
- Add a camera to the scene
- Adding lights
Testing:
- Light intensity, placement & colors
- Camera angles
Animation:
- Record the animation (see descriptions below)
- Rendering
First I would like to introduce Sn-can with this turntable animation. The model is posed in a dynamic pose to make it look more interesting.
Using the rig, I animated the legs from the side view. This way I could make the movement look extra mechanical and precise. With every step I made the top rotate a little, around it's local Y-axis, to simulate the impact of each step.
Using the rig, I animated the legs from the side view. This way I could make the movement look extra mechanical and precise. With every step I made the top rotate further, around it's local Y-axis, to simulate the impact of each step as Sn-can runs. The "head" is rotated forward to make it look like he is in a big hurry but his blasters are still at the ready.
Using the rig, I animated the legs from the side view. This way I could make the movement look extra mechanical and precise. When Sn-can plants his feet before firing his whole "head" rotates around the local Y-axis to simulate the seriousness of what is about to happen. With each shot the whole model gets moved back a bit due to the knock back. I decided to simulate be "bullets" with lights instead of 3D models so the animation won't get too messy.
Using the rig, I animated the legs from the side view. This way I could make the movement look extra mechanical and precise. When Sn-can plants his feet before firing his whole "head" rotates around the local Y-axis to simulate the seriousness of what is about to happen. The light intensifies as he charges his blast. Sn-can is moved back while he fires his blast due to the knock back and his blasters shoot up because of the blast and de blasters have to cool down.
For the damaging blasts I decided to go with different colors. For the first two blasts Sn-can rotates it's head and steps back. For the third blast he gets knocked under the chin and the whole model wobbles, shakes and powers down. It was a good exercise to animate lights.
SN-can - Front view
SN-can - Side view
SN-can - Top view
SN-can - Rig
For this project I really wanted to get some more experience in hardsurface modeling, storytelling, light and camera work and animation. I also really like low poly styles so I decided to expand my knowledge on that as well. While making a car is fun, I wanted to give it a little twist. So instead of wheels this car has thrusters. And instead of a smooth ride this car crashes at the end. I wanted to use this animation as an introduction to this character.
Using rock formations, trees and bushes, in different sizes and rotations, I was able to create an interesting background with a lot of variety. Using volumetric lighting I was able to creat a purple mist which gave the scene much more depth and made it look more interesting. I also wanted my animation to take place at night and I was able to accomplish that scene setting as well through volumetric lighting.
I made this animation using Blender 2.83. No sound yet but I will be adding some in the future.
Preparation:
- Collect reference materials
- Search for tutorials
Modelling:
- Build the car model
- Build all the rock formations using the Bool Tool
- Build the trees and bushes
Creating the Set:
- Add camera's to the scene
- Adding hdri world lighting
- Add mist and volumetric lighting
Testing:
- Light intensity
- Volumetric lighting
- Camera angles and flow
- Car animation
- Scene animations
Animation:
- Record the animation
- Rendering
ZAZ 968m
ZAZ 968m - Front view
ZAZ 968m - Side view
ZAZ 968m - Top view
All these objects will be used to fill the background and the floor of the scene. They will appear in different sizes and have different rotations so the scene will have a lot of variety.
First I made 3 different trees and 3 different bushes. All these trees and bushes will get scatered around the scene to create the look of a baren wasteland. Using weight painting the objects will get scatered evenly and the trees will be bigger where the intensity is higher.
First I made 8 different rocks. From these rock I build 5 bigger rock formations. All of these rocks and rock formations will get spread over the assigned areas using weight paint intensity. Where the intensity is higher, there will be more and bigger rocks and rock formations and vice versa. This wil give an element of randomness to the environment in the scene.
I felt there was a bigger story to be told after I finished my "Car Crash" animation so I started searching for ways to expand on it. I wanted more characters and an actual story to be told. That is why I came up with this new animation. It's about the old car, on it's way to it's last destination. Along the way the runs into some younger cars, also on their way in and in a big hurry, and crashes. The younger cars feel really bad and take him to a service station to have all the damages fixed. The next day the two younger cars wake up to the sound of honking. The oldtimer is totally fixed and as good as new, ready for an adventure. And so the trio ride off to greet the sunrise.
For this project I wanted work more on my low poly style and get more experience in animation, modelling and storytelling.
Act 1
-The desert is dark, night is falling and the sun is setting.-
SB1 – SB5*Camera 1, from the side*. So this old, rackety car (car1) races through the desert. Packed with luggage on it’s roof it moves forward -playing old jazz music-.
SB6 *Camera 2, from the front*. Now we see the car from the front with the luggage loosely packed on the roof, wobbling along as it races through the dark desert.
SB7 – SB9 *Camera 3, rear 3/4*. At first this car may seem like some old beat up car. Then we’ll switch to the back of the car to discover that this car doesn’t have wheels, instead it’s got thrusters. Rackety old thrusters with cables hanging loose. We’ll see it passing by from the side.
SB10 *Camera3, from rear Âľ to side to front Âľ*. We pan around the car until we reach an Âľ ish view of the old car moving through the desert.
SB11 *Camera 4, front ¾ bottom*. We see the thrusters and neat undercarriage of a second car. A shiny new Delorean DMC 12 (car2), the Back to the Future model. 4 Blue, clean thrusters at bottom and 2 big ones at it’s back. It moves much quicker than the first car, as can be seen through the environment.
SB12*Camera 5, 3/4/side top*. We see a third car racing through the dark desert. An old red Volkswagen T2 (car3) van. On it’s roof it’s got 2 surfboards attached, a huge speaker blasting -music- and a enormous red thruster, blasting the car forward. The T2 had got some rust here and there but overall it looks decent. The red van has 2/4 hover thrusters at it’s bottom.
SB13 – SB17 *Camera 3, front ¾ on car1*. We hear the horn of the car2 before it speeds past, on the left, car1. Car1 is startled and swerves to the right. Then we hear the horn from car3 as it passed car1 on the right. Car1 gets startled again and swerves to the left and quickly crashes into the rocks in front of it with a loud -noise-.
SB18 – SB20 *Camera 3, swerves around to see car2 and car3 from the back). Car2 and car3 hear the crash and break immediately. Car3 breaks hard and it’s back is lifted higher than it’s front before bumping back. Car2 breaks and makes a half circle so we see it from it’s side/ front ¾. After having stopped completely car2 first drives back towards car1. Car3 awkwardly turns around and follows.
SB21 *Camera 6, far off so all cars can be seen*. We see the 2 cars moving towards car1.
SB22 *Camera 3, back on car1*. We see both car2 and car3 stop near car1. Fade to black.
Act 2
-The desert is even darker now, night has fallen-
SB23 – SB24 *Camera 1, side and bottom car1*. Car1 thrusters don’t work anymore but he does seem to be flying. We move from car1 backwards to see cables attached to it’s back spoiler(?) which run up to the car3 spoiler. Car3 flies higher.
SB25 *Camera 3, Âľ front*. We see Car2 in front, flying. Car1 in the middle dangling and Car3 at the back also flying higher. You can see the damage to car1
SB26 *Camera 6, side*. We see all 3 cars moving forward in the dark desert. Fade to black
SB27 *Camera 7, pov car2*. First we see the car moving forward, a few seconds later we see an old gas station/car shop coming up on the right. Lights are still on.
SB28 *Camera 6, Âľ top view, cars and gas station are visible*. We see the 3 cars turning up on the gas station driveway. -Fade to black, repairing noises-
Act 3
-The sun is rising and the desert is waking up and looking more yellow orange-
SB29 *Camera 8, front on Car2 and car3*. We see car2 and car3 sitting in the parking lots, lights turned off as if asleep. Left front thruster of car1 appears in screen on right side. -Car1 honks- car2 and car3 “wake up”, the lights turn on and they come off the ground as their thrusters turn back on.
SB30 – SB31*Camera 2, front car1*. De view pans up towards a high side/ front3/4 view and we see that all damages to car1 have been repaired. In fact it’s got some of it’s rust and dents removed and it looks much better.
SB32 *Camera 8, front car2 and car3*. Both cars -honk- and the fronts of the cars lift up a bit more, as if happy.
SB33 *Camera 6, panned down a little*. We see all 2 cars leaving the lot. First Car2, Car1 and car3, we hear a -flick of music turning on-.
SB34 *Camera 8, on the desert road*. We see the 3 cars driving towards the sunrise.
-The end-
Concepting & Preparation:
- Collect reference materials
- Write the script
- Draw the storyboard
Modelling:
- Build the new cars
- Add particle systems
- Build the service station
Creating the Set:
- Add camera's to the scene
- Add new lighting for the new cars
Testing:
- Light intensity
- Volumetric lighting
- Camera angles and flow
- Car animations
- Scene animations
Animation:
- Record the animation
- Rendering
Editing:
- Add music and sound effects
For this car I wanted to make it look brand new. So there won't be any dents or rusty patches on this car. I was inspired by Back to the Future. I also wanted to run some tests using particle systems.
This VW has alsways been one of my favourite cars so I had to use it for this animation. This model is a combination of the ZAZ and the Delorean. It's got some rust here and there but not as much as the ZAZ. And it uses thrusters like the ZAZ but it's also got the particle system like the Delorean.
Younger car number 1. It's really shiny and brand new.
Younger car number 2. This one has some rust patches bit it's still got a long way to go.
The service station where the older car will get patched up. Inspired by the retro Shell gass station in Aberdeen Washington.
For my internship, and for one of my extra courses, they asked me to make some animations for their logo. While this logo is 2D I animated it in 3D. The company where I did one of my internships was looking for some logo animations for their website and as a loading screen icon for their product. After a brainstorming session with the creative team we decided I should animate a few of the ideas and make some variations of those ideas. I also made some sketches which I could use throughout this project. These animation were made using Blender 2.9 and they were redered using EEVEE.
Concepting & Preparation:
- Brainstorm session with the creative team
- Draw some sketches
Modelling:
- Model the logo
- Model the text
Creating the Set:
- Add a camera to the scene
- Add scene lighting and effect lighting
Testing:
- Timing of the logo
- Light intensity
- Word timing
Animation:
- Record the animation
- Rendering
Editing:
- Reformat to a webpage format
Made using Blender 2.82.
Made using Blender 2.82.