Live Brief

Week 01

Digital Matte Painting

Matte painting is a technique used in visual effects (VFX) to create detailed backgrounds or environments that enhance live-action footage. It involves painting or digitally creating images that are composited into scenes, allowing filmmakers to depict expansive or fantastical settings.

in a class we created a matte painting based on a given concept, like a sci-fi city or skyfall. I gathered various random images and used Photoshop to blend them into a cohesive artwork. This practice was really valuable, as it challenged me to generate ideas quickly and develop a concept and storyboard under time constraints. It was an excellent way to kickstart my creativity and explore new possibilities in visual storytelling

SIFI City 

 

Skyfall

 

Week 02

Frankenstein

We discussed about creating a scene from Frankenstein. Making a shot, environment, character, or anything else was the challenge for that. The exercise was about to create something that suited Guillermo del Toro’s style, and we were free to choose what we wanted to accomplish.

This exercise aimed to set the scene for Guillermo del Toro’s upcoming Frankenstein film. We were given two weeks for this, which is the most challenging aspect, but from an industrial perspective, it would be a fantastic practice.

Story Board

For this, I made a story board in which I attempted to provide a quick overview of my film, concentrating more on the surroundings and utilizing computer-generated imagery to create a scene in which a person is passing a cemetery and sees Frankenstein and run away.

 

After discussing about my idea , we came to the conclusion that practically it would be too much to achieve with 2 weeks of time, accordingly that I was advised to begin with a specific scene or scenario. In considering this, I made some changes to my project: I would be creating a scene in which Frankenstein is excavating a grave in a terrifying cemetery.

 

Week 03

Frankenstein – Work in progress

I have begun working on the idea that I came up with in the last lecture. For this assignment, I decided to use Unreal Engine. as I wasted to illustrate a spooky environment for the scene.

Basic Landscape

I started by making a simple landscape in order to get a sense of how I wanted to frame my scene. I then considered creating a day-night cycle. I spent quite a bit of time making the Day-Night cycle, which was difficult, but it was fundamental to learn the details. Even though I was just going to use the cycle’s night scene, it was still really helpful to change the lighting, which went well with the environment I wanted to show.

I choose the London actual time because this day-night cycle can function with it. The fact that you can adjust the surroundings in real time was unexpected.

Time Zone

Selecting Longitude and Latitude

 

After playing with it for a bit, I was able to tweak the nighttime so that I could have the ideal amount of light. I then began adding trees and plants, using megascan foliage for this. And downloaded some Sketchfab 3D assets. After placing them in the scene, they gradually began to provide the atmosphere I was looking for.

Foliage and 3D Asset

Additionally, I received a Frankenstein model from Sketchfab, which began telling a story on its own after I put it in the screen. The moon and the clouds were magnificent in a truly stunning way, and the cemetery and Frankenstein with its tree were looking good.

1st Look for the Establishing Shot

Week 04

We are expected to present nearly completed work this week. Speaking of my project, which I have been working on for the past two weeks, I have come to terms with the concept I intended to convey through my narrative. I’ve created four shots so far that will perfectly set the atmosphere for the Frankenstein Film while also preserving Guillermo del Toro’s aesthetic.

The challenge I encountered was the animation scene with the shovel. For that, I manually animated the crucial frames, which finally resulted in a really lovely outcome.

Following that, I received helpful feedback and suggestions from our professors and my peers about making  the ground smoke even more scary. To do this, I had to make volumetric fog, which I attached to a big cube and created the materials for, which looks fantastic and gives the impression that the smoke is interacting with the objects in the scene.

Decals

I created a surface with dirt and leaves on the floor using decals, which are 3D-looking textures in the form of 2D plane. The best thing about decals is that they can be adjusted to any size or shape.

 

Niagara

With Niagara’s help, I created falling leaves, which gave the scene’s opening shot a more dramatic feel. As there are many trees in the scene without leaves, the idea behind creating falling leaves was to represent the atmosphere of autumn and the Halloween period.

 

Volumetric Fog

One of the best parts of the entire assignment is volumetric fog. Making the necessary materials for it took a while, but the effort was worthwhile. I started by creating the fog in the atmosphere, which allowed the light to disperse very smoothly. Later, for the graveyard scene, as we discussed in class, I created the ground fog based on the feedback I received. The volumatric fog worked well, and I considered compositing it in Nuke, but I realized that wouldn’t be very effective because it wouldn’t react with the assets used in the scene. However, this fog completely complements the asset.

 

Tree Texture

The trees’ textures were not particularly challenging, but the model’s default texture was extremely glossy when I got them to the scene. To fix this, I downloaded a new wood material and replaced it with the other one, which suited well.

Cloth and Hair Texture

Similarly, the hair and clothing had a rubber-like appearance, so I replaced them with a different fabric and hair textures that I got from Quixel Bridge, maintaining the same base color while replacing the other values.

Animation

I found an animation of Frankenstein digging on Mixamo website, so I downloaded it to Unreal Engine and assigned it to the model. However, the shovel was not connecting properly, so I attached the shovel to one of Frankenstein’s hands and then manually animated it. The outcome was excellent.

Rough Draft of the renders

 

Week 05

Final Establishing Shot

 

Extra Shots

 

Frankenstein Final Establishing Shot

 

Week 06

Wall-E

The major goal of this project, which we have all agreed to do together, was to teach us how to collaborate with others. Since we are six, we all agreed to work as a single team to accomplish that.

Story Board

For our group project, I came up with the idea for a storyboard featuring Wall-E. The sequence begins with a first-person perspective, where the viewer feels as if they’re driving down a road surrounded by zombies. As the tension builds, the zombies begin attacking the car. Just when things seem dire, Wall-E appears and swoops in to save the day, effortlessly fending off the zombies. To wrap up the scene, Wall-E looks directly at the camera and gives a confident “yeah,” adding a playful and heroic touch to the moment.

 

Week 07

New Story Board and Concept Art

During today’s session, after discussing the project, we realized that the original idea I proposed would take longer to complete than the deadline permits. To stay on schedule, we decided to pivot to a new concept. One of our team members presented a strong idea: in this revised version, we will feature a 3D model of Wall-E in the foreground, with a building in the background that will be destroyed. He also created a detailed storyboard, outlining the different layers and elements of the scene to ensure a dynamic and well-executed shot.

Final Layout

Breakdown

Matte Painting

Houdini

Unreal Engine

Maya

After presenting our idea to the panel, they really liked it but suggested one change: they recommended removing Wall-E from the scene, as it needed to be inspired by the film, and the character wasn’t necessarily required for the shot. However, they loved the concept of the building collapsing, so we all agreed to focus on making that part as compelling as possible.

By the end of the session, we had fully aligned on the new direction and were ready to move forward. We then began researching how we could enhance the shot to match the quality of a feature film.

Another team member created concept art with the help of AI, producing several different designs. After discussing the options as a group, we collectively decided on one concept art for the CGI elements and another for the lighting of the shot. These final selections will guide our work moving forward and help us achieve the desired look and feel for the scene.

Concept Art for CGI Elements

Concept Art for Lighting

Once we finalized our approach, we began planning how to tackle the project based on our individual strengths and areas we wanted to improve. As a team of six, we divided the roles to align with the project requirements. One team member was assigned as the Effects Artist, responsible for the main building collapse using Houdini. Another member took on the role of 3D Modeling Artist, working with Maya and Blender. We also designated a Matte Painting Artist, who would handle the creation of backgrounds and textures using Photoshop and 2D compositing software. Additionally, one team member became the Environment Artist, working in Unreal Engine to bring the scene to life.

For the extra elements, one member volunteered to model a truck in Maya and later create the explosion effects to enhance the shot. This clear division of roles allowed us to efficiently collaborate and utilize everyone’s strengths.

Setting Up Camera and Position of the frame and 3D Assets

After deciding on our objectives, we received valuable advice from our professors. Since we would all be working in different software, they recommended that we pay special attention to the camera setup to ensure consistency across the project. In response, one team member quickly created a rough draft outlining the positions of the building and truck, and set up the CG camera in Maya. Once we had the camera data, we were able to move forward, with each team member working within their assigned software, aligned with the established camera position.

Position of the frame and 3D Assets

 

Week 08

Progress on the Environmental Part of the Project

As I was responsible for the environmental aspect of the project, my task was to ensure that the foreground of the shot maintained a sense of realism and effectively enhanced the overall experience for the audience. To begin, I started by positioning the blocks to form a basic building structure, creating a rough reference for the shot. This allowed me to establish the foundation for the scene and ensure the elements would work together seamlessly as the project progressed.

Positioning the Shot

 

After that, I set up the camera, ensuring that, before building the scene, it was crucial to establish the basics. Working on these foundational elements first was important, as overlooking them could lead to issues later on in the process.

Setting Camera

 

After setting up the camera, I began working on the floor. I decided to create my own textures for this, which took some time due to the numerous nodes I had to work with. My goal was to design a floor texture that appeared damaged, allowing me to add additional details like dirt and cracks, enhancing its realism. This process required careful attention to detail, but it was essential for achieving the desired effect.

Floor Texture

Making Layer Texture

Making Single Shade

Unfortunately, my initial approach didn’t yield the results I had hoped for, so I had to quickly reassess my plan. I decided to source textures from the Fab, and with these, I opted to use decals to give the floor an aged, damaged, and imperfect look. This approach allowed me to better meet the project’s requirements and achieve the desired effect.

Once the floor texture started looking appealing, I decided to create a road, especially since we had a truck in the scene. This addition would help advance the story and fill up the space. Furthermore, with a building on the side, the road would fit well into the overall composition.

After creating a basic 3D model of the road in Unreal Engine, I began adding the sidewalk and continued working on the road textures. As there weren’t many damaged road textures available, I chose one that clearly hinted at damage and the apocalyptic theme. I then used decals to make the road appear rough, dusty, and worn. This process helped build the atmosphere and contributed to shaping the vision for the project.

By that point, I had received the 3D models from the rest of the team. To get a sense of how everything would come together, I placed them in the scene as a draft, allowing us to check the scale and how the elements worked in the environment.

Floor textures and 3D Assets

I also imported some Megascans assets and foliage, such as trees, into the scene to check if the scale was appropriate and aligned with the overall composition. This helped ensure that all the elements worked well together and maintained a realistic sense of proportion. To further enhance the atmosphere and get the right feel, I added ExponentialHeightFog, and the result was incredible—it really brought the scene to life.

Adding ExponentialHeightFog

 

As everything was progressing smoothly, I began adding grass and plant foliage to the scene. This was an important part of the project, as it helped convey the idea that, despite the destruction, traces of life and the possibility of existence hadn’t been completely wiped out.

Adding Foliage

 

Week 09

Once we received the test render with all the assets created by the team, I realized that the camera values I had used were incorrect. After working through the issue and collaborating with the team, we were able to identify the problem and adjust the camera to match the scale that aligned with both the team’s camera settings and the ideal project scale we had established earlier.

Wrong Camera Scale (Before)

Correct Camera Scale (After)

 

After correcting the camera scale, I noticed that the shot became much wider than expected. To address this, I began adjusting the scale of the objects, starting with the floor, and then scaled everything up according to the exact data we were supposed to use for the project. The main challenge we faced was the discrepancies in measurements when using different software, but once we resolved that issue, everything fell into place and the scene came together as planned.

Project scale

Since the space around the building felt a bit too vast, we decided to create a two-lane road to help fill the area. Additionally, one of our team members created a 3D model of a car, streetlight, and trash bin to add more detail to the scene.

Expanding scene

After that, I worked on the asset created by one of our team members, who was focused on developing 3D assets for the project. She designed an incredible building for that purpose.

Rachel’s Building with Textures

 

Next, to give the building an abandoned look, I applied decals to make it appear as though it hadn’t been used or maintained for years. Everyone was pleased with how Rachel’s 3D model looked in the foreground, so we decided to leave it as it was on both sides.

Adding Decals

 

In the end, once all the 3D artists had finished texturing their parts of the project, I compiled everything together to see how it would all look. It turned out great, and that was the moment to render the final footage.

Truck Texture

 

At this point, everything was looking great, so it was time to render the scene. However, I removed the main building and the truck from the scene, as these two elements required a Houdini simulation. We decided to render them separately and then composite them later.

Final Environment

 

Final Environment

 

Final Wall-E 

 

Week 10

Christmas Advert

In today’s session, we received a brief to create a Christmas advert. We have some time to brainstorm and come up with three ideas, from which one will be selected to move forward with. Once the idea is chosen, we’ll have two weeks to work on the project.

For this, I came up with three different ideas. The first was to create a hot chocolate scene where the liquid drips, and marshmallows fall into it. The second idea was to design a snow globe with particles swirling inside. The third idea I presented was for a burger franchise, where sauce would drip out in an appealing way.

The intention behind these ideas was to help me learn the missing elements I want to improve my skills in. With the hot chocolate idea, I can learn how to work with fluids in Houdini. The snow globe concept would teach me about working with particles in Houdini, and the burger idea would allow me to explore thick fluids interacting with mesh objects in Houdini.

Presentation

Christmas Advert Pitch Presentation

Pitch Presentation

Pitch 01

Pitch 01 Approach

Pitch 02

Pitch 02 Approach

Pitch 03 Approach

 

Finalizing the Hot Chocolate 

For this project, my hot chocolate idea was selected. My plan is to create a 3D model of the cup and saucer, along with a fluid simulation in Houdini for the chocolate. I’ll also model the marshmallow in Houdini and set up a falling simulation for it.

Reference Image 01

 

Work in Progress

Maya 3D Modeling 

I started the project by modeling my hero object, which was a cup and saucer. It wasn’t particularly challenging as it was a straightforward hard surface modeling task. I didn’t spend too much time on it and was able to complete both the modeling and texturing within Maya.

Cup Saucer

 

Houdini Simulation Progress

After that, the challenging part began. I decided to create my hot chocolate simulation in Houdini, as it is highly sought after in the industry. This presented the perfect opportunity to learn the software and enhance my skills. To start, I imported my 3D model into Houdini.

Importing 3D Model in Houdini

To create the flip fluid simulation, I used the flip fluid node that is already available in the software.

Creating Flip Fluid Slover

Flip Fluid Solver

Simulation

Voxel Points

After trying everything, it was difficult to pinpoint the mistake I was making, as the simulation wasn’t progressing beyond a few frames.

 

2nd Trial

as the first approach didn’t work as expected. After doing some research online, I learned that I had too many points from the start, which was preventing the simulation from functioning properly and making it difficult to see the full result. This time, I adjusted the particle separation to improve the simulation.

Increased Particle Separations

 

Creating the mesh of the Fluid Particle Separations.

Mesh Fluid

It began to look like what I had planned in my storyboard. The emitter was releasing the fluid just right, giving it a thick, milky look. To make it more realistic, I changed the fluid color to brown. However, when it hit the collision object, it still leaked from the bottom. I spent a lot of time trying to fix this, and although the collision looked fine, the fluid kept leaking through small gaps. So, I decided to restart with a new setup for the emitter to keep things simpler and netural.

3rd Trial

Emitter

 

Progress

Dop Network

 

 

Week11

Deleted Faces to make Cup Fluid

Changing the approach to cyclinder for the cup fluid.

Using tube as cyclinder to make Cup Fluid

Marshmellow

 

Simulation

Simulation

Fail Velocity

 

Collision

 

Fluid Collision

 

 

 

Week 12

 

Final_Advert_Presentation.pptx

Final Presentation

1

2

3

4

5