
Critical Reflection
Miro Board: https://miro.com/app/board/uXjVM4xIshw=/
Schedule: I have established a basic schedule and timeline for the project, identifying which tasks need to be completed within certain time periods in order to meet the submission deadline. This has also enabled me to plan the project distribution in greater depth.
As this was only a rough block-out of the timeline, I expected it to evolve over time. However, it allowed me to have a general look at the scope of the project and what I needed to accomplish at each stage. It was a useful tool to motivate me to work and keep track of each task (including reflective notes for the blog).
Moodboards: I found detailed references for the assets and room structure, established the overall mood of the scenery, decided on a colour palette, and brainstormed VFX ideas.
It seems that I may have gone overboard with my brainstorming, as I fear I may have included an excess of details. While I do believe that these details enhance the overall worldbuilding and visual appeal of the setting, unfortunately, it was not feasible to execute all of my plans within the allotted timeframe.
Modelling: I developed the concept further by using the technique of kitbashing and creating multiple block-outs. My focus was on creating assets that would help me reinforce the particular style of the room and add visually interesting details to keep the audience engaged. Finally, I added the necessary Art Nouveau details by outsourcing the models.
During the project, I devoted a considerable amount of time to creating block-outs for the room, which led to a more captivating area to explore. In retrospect, I am satisfied with the outcome but recognize that I may have spent too much time on this phase. I could have transitioned more promptly to the asset modelling stage, which proved to be the most challenging aspect of the project. To expedite the process and shift the focus to texturing and lighting, I decided to outsource some models. However, this approach had its limitations. Some of the models had incorrect topology, which necessitated additional time for correction. Additionally, certain assets were not imported properly into UE5. Also, at times, downloaded models disrupted the existing style of the room.
Texturing: I corrected the UVs of the models and tested them in Substance Painter. After selecting one of the simpler-looking models, I applied procedural textures to it and created multiple templates for the remaining assets. Then, I textured each asset individually in the scene, modifying the templates (smart materials) to suit each model’s specific details and purpose, and ensuring that their colours complemented each other.
I am very pleased with the texture, especially when viewed up close. The subtle brush strokes complement the stylized look of the project, and the mix of colours and layer styles in Substance Painter matches the hand-painted textures created in Photoshop. Adding the outline in Unreal Engine 5 helped tie everything together, and it is my favourite part of the project, next to lighting. If I had more time, I would keep more details of the Photoshop patterns and blur them less. However, I am overall happy with the balance between detailed and generic textures that I managed to achieve. With so many assets in the scene, it was important for me to not overdo the painterly textures, so the scenery wouldn’t be too chaotic.
Lighting and Rendering: Initially, I set up the basic lighting for the scene and then proceeded to add all the models and textures. During the importing, I encountered a few issues with overlapping UVs and wrong polygon counts on some of the models. However, the overall process wasn’t too complicated. Unfortunately, I had to redo my Blender effects in UE5, which caused my scene to lose some of its unique stylized look. I feel that the blueprint Niagara effects look more generic. Additionally, I had to rush through the animation process, which lowered the quality of my final product.
In hindsight, I realize that I may have benefited from devoting more time to creating various camera shots, incorporating additional post-process effects, and shooting more scenes than necessary to assess their compatibility with one another. I acknowledge that I missed opportunities to experiment with more imaginative camera angles and capture more interior shots, which ultimately resulted in the video feeling incomplete and hastily done. While I am content with the life that my Niagara and Houdini effects breathed into the project, I now recognize that I could have intensified the visual intrigue with more animated elements, such as moving plants and cats.

Final Thoughts
Overall, I am happy with the outcome of my Final Major Project. I had the opportunity to experiment with Houdini and photobashing techniques, which I found to be very useful and plan to continue using. I successfully achieved my primary goals for the project, which included creating a UE5 scene, generating stylized procedural textures, and designing compelling lighting simulations. During the project, I learned a lot about different lighting approaches and hand-painting materials. Creating more mock-up lighting scenarios with free Unreal marketplace assets would be a good practice and addition to my portfolio. As for textures, I am interested in coming up with some organic shapes like people and animals, emphasizing details through hand-painted materials.
I regret not being able to focus on compositing and post-production for my project. Despite following a well-planned schedule, I encountered some unexpected setbacks. As I had mentioned earlier, during the initial stages of my work, I spent a significant amount of time exploring various styles and creating block-outs. While this allowed me to create a more detailed scene, it also caused me to fall behind on my original timeline. I plan to revisit it and add more visual effects to enhance it. I have started to rerender some scenes with passes to recreate the accurate UE5 lighting. Additionally, I believe that learning about compositing lighting will be beneficial for my career. After reviewing the finished product, I feel it lacks movement and needs more dynamic elements, such as subtle plant movements with the Niagara simulation around them. Also, I regret using a low poly model for the cat, and I think that adding more lively and animated elements will positively transform the scene and make it more immersive.
I would also like to explore Unreal Engine 5 more as a rendering tool. I have started a new stylized project, focusing on lighting and rendering in Unreal Engine, using only outsourced models. I will upload all of my future work to the Artstation: https://cervesna.artstation.com/.
I believe I have built a solid foundation for my future projects and have identified the mistakes I tend to make. My goal is to further enhance my skills in lighting and rendering, and hopefully secure a job that allows me to focus on these particular skills.