Week Seven
- Feb 23
- 6 min read
Updated: Feb 24
Blog posts arranged newest to oldest.
To-do:
Refine flower vellum simulation
Refine shot 3 camera and ring animation
Look into adding the frost simulations after completing the flower simulation
Mentor Feedback:
Is it between paper or real flower? Can’t tell right now, feels like paper - Flower and leaves are missing SSS, everything looks very flat even the environment feels flat
Overall colors are a lot better
The defrosting can be more organic, right now it is very obviously defrosting from the y axis
Purple silkiness is very beautiful - but make the color a little more punchier
Light bulb on ring in water shot is missing?
Water shot- sky feels stormy- disconnect between water and sky
Don’t like the background for shot 5 - Backgrounds in Oura ring branding are a lot richer
In flower shot, out of focus light shaft on the background
The light should pick up the rim of the ring and it should be the brightest part of each shot
The frost is too white, right now it is overpowering the image- the white of the ring should be the focus
Maybe studio lighting in flower shot, maybe an actual environment will help this shot, the light that is affecting the flower should hit the backgrounds - Think about how you would shoot flower shot in real life
Too many little things in the middle of the flower, the ring is getting lost in it - Maybe not have them at all? Or have them blend into the pink a little bit - Maybe smaller petals in the center of flower
Week Seven Submitted Video:
2/21/26
To get a more accurate animation, I spent time with the help of my teammate Cae, and figured out how to use vellum grain with the shapematch constraint to simulate the flower and the ring all together in the same vellum simulation.
I then played around with this way of simulating and captured multiple different versions that I will be using pieces of or using as reference for the ring animation.
My biggest issue with simulating the ring is that realistically, it would fall lower into the petals and out of camera view and realistically it would not fall out unless the flower almost completely bent over. This is ok for the ad as most of Oura ads include the ring moving in a sentient way, but this means I have to animate most of the motion.
If I were to use the simulated motion, here is the result.
While this may be realistic, I believe it makes the ring look too jerky and is not the gentle vibe that the ad is going for.
If I completely hand animate the ring, I personally struggle a lot with keyframe animation, as it is not my strong suit.
After working on the animation for so long, I believe I have serious blindness with it and need to take a step back. Because of the camera and ring animation as well, I had to change the stem movement and timing of the first bounce as realistically, the ring would fall out on the first bounce. This changed the movement I spent all week working on a lot and is something I will be looking back into right away for next week.
2/20/26
After feedback from the group and from professors, I continued refining the motion of the flower, focusing on the secondary bounce that would happen if the flower did fall over and bounce back up.
With the simulation on the flower petals, here is what the motion is looking like today.
I also wanted to focus on adding more variation to the petals to have for a more realistic look. This week, the flower model is actually going back to being paper thin rather than having any thickness. This also means that we are able to use opacity maps again but I do not want to rely on those.
I focused on adding noise to the petals both before and after the simulation. Some of the petals are based on growth while some noise effects all of the petals equally.
before sim:
I am also creating variation in noise by separating out groups of petals and applying different noises to those groups of petals.
I am also doing this after the simulation as well.
2/18/26
Combining the new flower simulation, leaf simulation, and stem simulation together, here is my progress for today:
While it was something I could not achieve last week, I began this week by focusing on how I could simulate the leaves, rather than having to animate them. I hooked the model up to a very similar vellum cloth simulation that I used for the flower.
At first, I had a lot of issues with the geometry but with trial and error, I was able to get the model simulated.
Next, I began remodeling the flower. I wanted to push the amount of petals even further this week and also rework the inside of the flower to get rid of the stamen. As per professor request, I picked one singular reference photo and very specifically modeled the shape after that image.
For the inner part of the flower, I ended up building a completely different mini flower that I send inside the bigger flower. This gave me more control as well over how much the inside would bloom versus the outside as the ring will be sitting on the middle, so it will need to stay more stagnant.
Here is the new flower model after really trying to push variation in the petals formation and get and a more realistic shape. Last week the model had 35 petals and this week is has 85.
The biggest critique this week about the flower was that the motion felt way too stiff. Therefore, I started adjusting parameters in my vellum simulation to give it a much more flowy feel. After simulating lots of tests, here is how the flower is looking.
Another huge critique was the way that the stem was staying stagnant during the shot as well. To create a more realistic stem look and motion, I actually ended up using a tool that my peer, Hannah Kim, is using for her own group's project.
After loading this tool into Houdini, I was able to use the trunk maker node and the wind motion node to get a more realistic movement from the stem, even with no outside force on it but rather just wind.
Here is the motion I have achieved for the stem using this method so far.
2/17/26
Mentor feedback
Molly
Comp vector aovs and then motion blur in comp
Hailey
Softening alpha, smearing color to the edge, eroding alpha also works
Beck
Is it between paper or real flower? Can’t tell right now, feels like paper
Overall colors are a lot better
Flower- add SSS, sim also should be more realistic, combination of lookdev and sim
The defrosting can be more organic, rn it is very obviously defrosting from the y axis
Hailey
Purple silkiness is very beautiful
Flower and leaves are missing SSS, everything looks very flat
Even the environment feels flat
Light bulb on ring in water shot is missing?
Water shot- sky feels stormy- disconnect between water and sky
Don’t like the background for shot 5
Kyle
Backgrounds in Oura ring branding are a lot richer
Richer gradients
In flower shot, out of focus light shaft on the background
Even flat color background looks interesting in the original oura ring reference
The light should pick up the rim of the ring and it should be the brightest part of each shot
First shot, more earth tones in the background
The frost is too white, right now it is overpowering the image- the white of the ring should be the focus
Flower shot, background is missing some shape
Cloth shot, make the color a little more punchier
Motion graphics looks great
Beck
Maybe studio lighting in flower shot, maybe an actual environment will help this shot, the light that is affecting the flower should hit the backgrounds
Kyle
Think about how you would shoot flower shot in real life
Molly
Too many little things in the middle of the flower, the ring is getting lost in it
Maybe not have them at all? Or have them blend into the pink a little bit
Beck
Maybe smaller petals in the center of flower
Personal To-Dos this week:
Focus on just the flower simulation, leave the frost for now and see if I can come back to it later
Focus on getting the motion of the flower to be more realistic
Rework the shape of the flower to avoid the stamen and be more realistic.
After feedback from class today, I began to really focus on finding more references and have combined them all in a presentation for myself.
