Tumgik
idp-dome · 3 months
Text
Final Stages
This is the description of the very last stages of our interdisciplinary project in which each one of us contributed within their own skills.
During the time we each spent developing our separate scenes, we kept communicating as a team to make sure the final video outputs ended up merging together as a puzzle, and no huge differences in style could be noticed throughout. Some things we discussed are aspect ratio, video resolution and transitions. We encountered some rendering issues as well since we are new to this type of visualisation that is made for VR and 360 domes, but we made sure to keep contacting each other with updates on the matter and finding fixes for the rendering issues. Talha was especially helpful as he had more experience than Abhinav and I in both Unreal Engine and Blender, so he was able to guide us. We were eager to test our creations on the last Friday before the submission, but the testing ended up getting postponed to the day of the exhibition which made us more worried due to the lack of time to fix things in case we encountered issues with the projection.
After we were done with rendering our footage we sent all sequences to Anosh who was responsible of compositing, editing and adding sound to our project. He made sure to do the proper research before starting his practical work, which was great, as he ended up coming across useful techniques for working in After Effects. At some point I was contacted regarding something that did not seem quite right in the final output of the video: it turned out that I had set up my camera in a different way compared to Talha’s. In fact, I had put my camera in the scene in a way that it’s positioned beneath the particles that I wanted to have projected at the top of the dome, which was clearly noticeable based on the full-dome render output. In fact, the particle systems I had rendered ended up being on one side of the circle rather than the center, for this reason, I decided to contact one of our classmates who told me that he had done it the same way as I had. Since we were not sure if we were supposed to direct the camera in a way that faced the visuals or not, I suggested to clear out any probable issues. I suggested we create two versions with two different scenarios that we’d test the next day, but Talha raised some valid concerns regarding rendering time. Fortunately, Anosh started doing some research to find a solution, and he came up with a way to flip the 360 rotations of our camera angles in the panoramic render sequences in After Effects, which granted us more flexibility.
Unfortunately, though, when the day came we were not able to see our project come to life when the event was cancelled as the university was not able to provide us with the dome installation needed. Nonetheless, we were very satisfied with the final result as we watched the video that we produced. It was nice to test our abilities to collaborate and create something that is both visually pleasing and satisfying.
0 notes
idp-dome · 3 months
Text
Summary (Talha)
At the start of the project, we had some ideas that we thought were good enough to work with this assignment. As Joe mentioned in his blog post. We wanted to keep them a plan B in case we decided to shift from our original concept. I am copying those ideas here from Joe’s post.
Journey through desolate wastes (low-end soundscape, wind that ends up picking up as we ascend a mountain, Journey game)
Floating through space (slowly revealing some sort of hazard in space, maybe a wormhole)
Paralysed patient in lab or hospital (geometrical dimension)
Maybe combine all ideas through flashes. Basically, Walking Simulator without the walking.
A jungle with mysterious creatures around.
The problem we faced early on was we didn’t really know what type of dome projection we were working with, so the ideas were all over the place. If we had full information at the start of the module, we could have a clearer picture on what we wanted to do for this assignment. After, our first feedback session with Ian, we came to know that it’s best to not have sky or ground themed idea as it could be distorted in the projection or won’t have aligned properly in panoramic render, so we had to scratch most of our initial ideas since they include either a ground or sky or both. Back to the drawing board we went, and we decided to keep working with spatial environments while keeping our initial idea of moving through different environments.  
In summary, over this module we learned to work as a group and categorise work while considering each other’s skills. Our aim was to produce a finished artefact before 9th February to present on the dome, learn about the immersive video workflows during the creation and focus on team research, team skills and collaboration. Every member of the group worked to the best of the abilities for this assignment. In the end we were able to achieve our goal that we set at the start of the assignment. We had multiple meetings; we were vocal on each other ideas and feedback and communicated regularly. All in all, I am happy with the work we did as a group.
Submitted By: Muhammad Talha Farooq
0 notes
idp-dome · 3 months
Text
Completion Phase
In the final parts of our project as a team we worked mostly from Talha’s portal test file from last week. Using this as a reference, the team created their own versions of a black hole/warp visual. We also faced setbacks regarding not being able to test on the dome Friday nor Monday. At this point there was not much more to update on, the collective works were being polished by their creators. We tried to source license-free music that could compliment the visuals we had created. As a concept artist, I personally had little to contribute at this stage so focused mostly on gathering examples of our work and collecting it here. At this stage we mostly communicated via our project-discord and whatsapp group. Support was high and every member of the team was vocal and responsive. As there is not much else to update until we get to test the project in the dome I will summarise some of the challenges we faced. Firstly we had very little working time to try to organise our team and it’s working relationship, generate an idea and execute it effectively. While our team was motivated and responsive we had no issue organising ourselves and finding our roles, however we did spend at least two weeks just generating ideas and anytime we had a functional idea we were blindsided either by limitations of the project and theme (no Horror, no Trance music, the unknown shape and structure of the dome that left many questions irritatingly ambiguous, etc.) So it felt like we were running uphill every step of the way only to get to the top and have another hill revealed to us. While this is part of any creative project and adapting and overcoming is a useful skill to cultivate, this was very frustrating as it was a recurring issue and I feel if the parameters were laid out more effectively we wouldn’t have had to waste so much time. Additionally the dome was unavailable for testing on the assigned days so we were left with a lot of uncertainty going into the final stages. Also as a team we were trying to leave room for everyone to work the skills they wanted to, which led to some of us stepping back from duties they were more suited to, to allow someone else to cultivate desired skills. This hampered our effectiveness a little, especially early into the project, but we recovered effectively and efficiently, largely due to our excellent communication. So to sum up, our team managed to overcome many challenges and limitations to deliver an effective product that sadly does not fully reflect our abilities and given more time I believe we could have worked more effectively and definitively as a team.
0 notes
idp-dome · 3 months
Text
Meeting Notes
This is a collection of notes I took to organise my team throughout meetings:
25/01/24 Meeting 1 Group 1 - Anosh Ahmad - Video Editor, Animator, VFX Artist - Wants to learn more about Unreal Wordpress Trello After Effects Audio Design/SFX Concepting Joe Abdel Sater - Props - Environment Artist, Modeling in 3DS-Max and texturing in Substance. - New to Unreal Texture Artist Lighting Artist Eric Bergner - Concept Artist, 2D Artist, experience in modeling with Maya as well as animation. - Art Deco High Contrast Storyboarding Concept Art Patterns, etc. Muhammad Talha Farooq - Environment Art, 3D props- modeling in Blender, texture in substance, Unreal engine. - Source Control setup Unreal Modeling Abhinav 3D Modeling Auto Mobile experience with particles - Particles Modelling Ideas Door Maze Immersive Environment Character focused Mirror Dimension Cube Portal - Speaker as a character - Glados? Team Fortress 2 Announcer Voice/AI voice - Anosh Story Anosh - Control Video 04:00 - 04:30 ‘Story’ Core Directing user with sound The maze is a malicious character, we don’t need a guide character. Audience is the protagonist. Get the character across through sound design. -Start- Character/viewer blinking into consciousness. Get out projected onto wall Boring empty room, clock tick tocking sound, builds up then cuts out, knock on the door, drawing the audience’s attention, then door opens and creaks open very slowly (can already eat up 15 to 30 sec. The props from the opening room can be used as visual motifs that recur throughout. Blinking idea as scene transition Scribbles on the walls from people having tried and failed to get out before. Jobs Mohammad Setting up Unreal Joe Collecting Inspiration Anosh Collecting Inspiration Setting up Wordpress & Trello Abhinav ? Eric Concepting Next meetup Sunday 28th 29/01/24 Meeting 2 Joe introduced some major points regarding the shape of the Dome and whether it sat flush to the ground or had non-curved walls that it sat on top of. https://dl.acm.org/doi/abs/10.1145/2994310.2994339?casa_token=90RMeaIOeZMAAAAA:IMvGSGwdIaPBpJPXuRdd-XoYUCGRDCdJvTzLKrPR_wl3_6gMgDo2Z6Hav-TON2yqNM2DeXGAxAxqDA Talha introduced https://lumenandforge.com/everything-you-need-to-know-about-360-domes/ Talha set up an Unreal scene. Ideas for motion sickness prevention discussed. Discussing version control. Talha suggested simple fades for environment transitions. Can use strong audio design to build the transitions. Joe suggested it could create dissonance in the viewer. Anosh set up blog as communal blog on Tumblr: https://www.tumblr.com/idp-dome/740782994482233345/group-one-assemble?source=share Colourful, vivid How to use lighting to our advantage. Abstract Shapes and Landscapes. Questions: Dome setup specifics (walled or non-walled) What does he consider horror? Alternate ideas Journey through desolate wastes (low end sound-scape, wind that ends up picking up as we ascend a mountain, Journey game) Floating through space (slowly revealing some sort of hazard in space maybe wormhole) Paralysed patient in lab or hospital (geometrical dimension) Maybe combine all ideas through flashes. Basically walking simulator without the walking. Focus on “best idea” and move out from there. Focus on experiences and sound design. ——— Everybody downloading Github in preparation for unifying version control. Jobs: Joe Setting up Lighting Talha Set up environment and movement through it. Look into rendering issue. Setting up second backup project to take functionality from. Eric Concepting visuals and some lighting and plot ideas. Concepts for Wednesday! Make Moodboard. Anosh Starting Audio work Collecting References for abstract environments possibly adding narration. Building Trello board Next meetup Wednesday 31st Jan
0 notes
idp-dome · 3 months
Text
Reflective Evaluation - Eric Bergner
We would meet once weekly to develop our creative concept through group discussion on Discord. We were very careful to include every member of the group in every decision. We started by defining everyone’s skill-set as well as what skills they would like to develop, even if they were outside of their expertise. We discussed ideas and approached them from multiple angels, making sure everyone was on board with it’s direction. Most of these ideas were stress-tested and made obsolete by limitations of the dome or theme. We continued this process until we eventually landed on an irrefutable idea. And while this idea felt like a compromise rather than what we wanted to do, the fact that we were not broken by our continued failures is an achievement in and of itself. As a group communication was our greatest strength. Everyone contributed with videos, images and discussion and no one seemed afraid to speak up and if they had not spoken in a while we always made sure, as a team, to get their side and discuss it until everyone was happy and on board. For my own work I feel I did in fact not contribute my skills in an effective manner. Due to the constant changing of our concept it was very difficult for me as a concept artist to produce work that was of benefit to the team. I’d created Moodboards, concepts and storyboards that were out of date almost the moment they were completed, so I took on more of a administrative role. Guiding the team, making sure everybody’s voice was heard and keeping notes of meetings. However as the work became more focused on communication among the modelers I took a step back from this as I felt like having a third party involved was simply slowing the transfer of information between the involved parties. I was also very instrumental in the generation of ideas for the project, something I absolutely love to do and the rest of the group were very receptive to my ideas, even if we had to change them due to limitations in how our work could be presented/projected. I concepted some ideas for transitions and characters through my storyboard and generated different environments and palettes through my concept art, however, as mentioned before we ended up pivoting away from these ideas later in production. I believe with a longer time-frame for pre-production I could have been more effective and I believe our final idea was more of a compromise than a fully generated idea. But nonetheless I am proud of my team and have learned a valuable lesson in how I can be an effective member of my team even when not directly involved as a concept artist and I believe that my inclusive and idea-focused approach was beneficial to my team overall.
0 notes
idp-dome · 3 months
Text
Production Research
My research during the production phase was mainly focused on creating spatial environments and abstract visuals. During the pre-production phase we had some meetings so we could finalize the direction we were going to take with this project. Joe come across a very useful article (https://doi.org/10.1145/2994310.2994339) that helped me understand the approach I was going to take with this assignment. The article was based on interactive dome experiences from user’s prespective and some technical aspects of creating said experience. I also came across this article https://lumenandforge.com/everything-you-need-to-know-about-360-domes/ which helped me understand how 360 videos worked.
During my production week, I tested with different scenes in unreal engine and researched about the panoramic capture tools that are built in Unreal engine 5. I used this documentation from Unreal engine official page https://docs.unrealengine.com/5.0/en-US/panoramic-capture-tool-in-unreal-engine/ that helped me immensely for testing different settings and see what works best regarding the panoramic captures within Unreal Engine. There were multiple times when I would get distorted or grainy images and I went back to this document to figure out what settings to change to fix those issues. Although, there were also a bunch of videos I watched on this topic and came across multiple reddit and unreal engine forums looking for a specific fix for an issue or asking others about the problems I was facing. I also learned that panoramic worked best when there is little to no ‘rotation’ on the camera and when there is little movement of the camera, otherwise the panoramic renders would be distorted and wobbly. I also helped other classmates with the same issues they were facing regarding the movement and rotation of the camera. At the end, I learned about some best practices tips and tricks when approaching panoramic renders and I can confidently say I would be using them whenever I had to work on scenes with panoramic setting.
While working within blender was fairly simple, I researched about the geometry nodes in blender and how one can create geometric shapes using geometry nodes for a non-destructive workflow while using very few resources. These short videos from blender studio helped me a lot while working with geometry nodes https://studio.blender.org/training/geometry-nodes-from-scratch/. For rendering panoramic images, I tested with different resolutions and settings like camera angles to see what worked best. At the end I ended up with 4096*2048 (4k) resolution with 240 frames and 30fps with each frame rendered with 200 samples. It took ~3 hours render time per video, and I created 3 videos so a total of ~10 hours went into rendering the image sequences out of blender.
Overall, my research during the pre-production and production was quite helpful in figuring out stuff and fixing problems and I really appreciate efforts of my team members and I am quite happy with the final output we got as a team.
Submitted By: Muhammad Talha Farooq
0 notes
idp-dome · 3 months
Text
Production Research (Abhinav Rshikesan Nair)
We discussed many concepts including realistic environments and after some meetings and consultations with our professors, we have agreed to work on abstract animations for the dome project. Also I believed it’ll be helpful for me to learn something new. As I was new to the abstract animation I had to research very well about that. Abstract animations are like a non-relative films without any acting or an attempt to reference reality or concrete subjects.
At first I was worried on how to start and then my team mate Talha gave some suggestions like some 3d animators in youtube such as Neb Motion (https://www.youtube.com/@NebMotion) and Ducky 3D (https://www.youtube.com/@TheDucky3D). They guided me on how to create abstract animations using geometric shapes in blender. I have shared my themes with my team mates for approval to start and they were supportive and gave confidence to me. I did research on the works of some of the famous abstract animators like Norman Mclaren, Oskar Fischinger, Len Lye and so on.
The main theme i had was to create an animation with rhythm and movement which look like a mirror dimension. So I checked some of the youtube tutorials by Neb Motion and it helped me to create the abstract shapes then I made my own animations, camera movement and after adding gloom effects with color made the animation look much more good similar to an illusion. After completing the animation the next thing for me to learn is how to render in panoramic view or as in 360 view. In that, Talha guided me on changing the settings for the cycles render in blender and after that I was able to render it. I rendered the animation in panoramic view as 4K (4096*2048) JPEG image sequence with 30 fps.
Anosh will work on these image sequences are later combined with the other animations done by Joe & Talha and make it into single video with hemispherical view with the help of AfterEeffects or Premiere pro. And Anosh will also work with the sounds to create the audience more involved.
I was pretty happy while working in a team, I think the first activity session which Ian planned helped our group to make a beautiful bond and that made us easy while working our project. I believe that the final result in the dome projector will give an amazing immersive experience to the audience.
Submitted by Abhinav Rshikesan Nair
0 notes
idp-dome · 3 months
Text
Follow up post
This is a follow-up post aiming to mention some additional information regarding the key steps that were taken during the production stage.
It was during our feedback session on Wednesday the 31st of January that we presented our final idea to Ian and Neil. We had previously explained to them that we wanted to create abstract environments/spaces with the main goal of creating deep immersion through audio-visual effects. To begin, we showed the instructors some of the reference images and videos we were using as inspiration as well as the concept sketches done by Eric. We then presented our methodology and the workflow we chose to adopt, which was later approved by them. During our discussion with Neil, it came to our attention that we had to use royalty-free music for our soundtrack, which made us realise that the music tracks we were thinking of using were no longer convenient as they had copyright properties. After the session, we had a team meeting where we agreed that as a start, Talha, Abhinav, and I will be working on creating around 40 seconds of footage in our own software of choice, which will be later rendered as panoramic images. Talha began by creating transitions for our scenes, I started by experimenting with abstract particle systems, and Abhinav learned how to create animated jellyfish that we could potentially use later on.
On Wednesday 7th of February, we already had some rendered 360 videos to show to Ian and get his insight. We had many doubts regarding the initial structure and technology of the dome projection so we made sure to raise these concerns during our session. It turned out that the angular positioning of people experiencing the projection would be arbitrary, and the only thing that we can be sure of is that people will be lying on the floor looking towards the top of the dome. For this reason, we decided to have all scenes symmetrical on the x and y axes, that way no matter which side the person is on, they will be having the same experience as others. Following the previous questions, we had regarding the shape of the dome; we did our own research online and looked at dome technologies posted on the websites of 360 projection companies. Ian later helped us confirm that the dome is indeed semi-spherical with the projection stretching from the top all the way to the bottom edge touching the floor. This meant that we had to focus on creating visuals for the top 75% of the interior surface of the dome instead of things closer to the ground as they won't really be seen by viewers. It was also advised to us that we do not put any main elements at the center of the dome top as there's a possibility they will get warped due to the low resolution of the projection in that area.
Our personal team meeting that followed helped us figure out how we'd like to continue working on our own separate scenes while keeping in mind the importance of seamless transitions between them.
0 notes
idp-dome · 3 months
Text
Project 2
In the start we were discussing on many concepts, such as creating a realistic environment with creatures or pleasant environment with lights & butterflies, animating abstract shapes. And then explaining those concepts to our professors they were impressed and gave some suggestions and guidances on what should  we need to focus on the dome. So we all agreed to proceed with abstract shapes.
We had team meetings on zoom and discord to share our ideas and for giving the roles for each. We divided the roles in the basis of 3 main roles, Pre-productin, Production and Post-production. Me, Talha and Joe were was part of production team where we,ll be creating abstract animations in 3D. Eric gave concepts and ideas with storyboards. Joe made an animation in unreal and I chose blender, and this was my first time working on abstract shapes, so I had to do some research for that in youtube and watched some tutorials. While researching I got inspired with works done by one of the youtuber named 'Neb Motion'. I shared those with my team mates for suggestions and they were happy with my ideas. I animated 2 abstract loops in blender combining geometric shapes with camera movements. Adding gloom effects and colors made it look more good like an illusion. On the last session with our professors (Ian & Steve) we explained our progress abd process by showing a rough render outs. They were pretty impressed with the progress. I was confused in rendering as panoramic, and my team mate 'Talha' giuded me on how change the settings for rendering as panoramic. I rendered those 2 animations in 4K (4096*2048) JPEG image sequences in 30fps and handed it over to Anosh for Post-production. He'll combine the all render outs given by us and i'll add suitable sounds to get better experience for audience. 
In overall, we are prepared for tomorrow, and I believe that the finalised look in dome projector 'll be amazing, however we are bit nervous on the working of projector. Anyway we have to do a trail before final presentation.
Submitted by Abhinav Rshikesan Nair
0 notes
idp-dome · 3 months
Text
Production
After our follow-up meetings on the 1st and 5th of February, we were pretty clear on what we were doing for this assignment. Pre-production was immensely helpful for this assignment. It helped us to experiment with different things see what might work and iron out all the problems regarding the technical aspect of the production phase.
The session on Wednesday 7th February was a feedback session, we created a space warp wormhole in Unreal engine as a pre-visualization and Eric did some concept art, we were able to present our final idea to Sir Ian and discussed our thought process and he was happy with the direction we were going to take with the project.
Since we used the space warp that I created in Unreal engine as reference, during the following week, we researched and experimented with different things that we could use. Since we agreed to do wormholes that would take the user to different dimensions, I came across abstract 'Video loops'. These are essentially visual sequences that run for 8,10 seconds they are made to run in a loop. Since Joe suggested that we shouldn't have a sky-ground and instead focus on spatial spaces, video loops were perfect for us and we could use those as dome projections. I am adding a video as a reference https://youtu.be/wViaZrGMzoI?si=-xa_GGEf_mXWV-Qh.
During my testing in the production phase, I made some basic scenes in Blender, and for non-destructive workflow, I used geometry nodes and modifiers to quickly make some abstract shapes and quickly iterate on those shapes. I followed some tutorials on YouTube to get an idea for creating perfect loops. I made some initial panoramic renders to test them in VR to check if they would work as 360 images. This video was really helpful in creating panoramic renders in Blender https://youtu.be/1qkldqKMvuM?si=1YtN3WCyJh6XzTKd.
For this project, I created 3 video loops with different abstract shapes and ideas that I thought would be visually appealing and colorful. Although rendering the final videos took most of my time I rendered all the videos as 4k JPEG image sequences in a panoramic setting. I handed all the image sequences to Anosh for post-production since he is doing sound design and editing to create a final video.
Overall, I am pretty happy with the current state of this project and I believe that in the end, we did well as a group and I am excited to test our final video and presenting it inside the dome.
Current Status:
I sent all the videos to Anosh for Post-production
Anosh working on combining all the videos in After Effects/Premier Pro, creating sound design, and exporting the final video.
Joe worked on creating a video using particles in Unreal Engine
Abhinav handed all his work to Anosh
Eric finished his concept work/storyboarding
Submitted by Muhammad Talha Farooq
0 notes
idp-dome · 3 months
Text
Bibliography
Talha Farooq
Blender Studio (2020) Geometry nodes from scratch, Blender Studio. Available at: https://studio.blender.org/training/geometry-nodes-from-scratch/ (Accessed: 05 February 2024).
Brandon’s Drawings (2022) Create 360 panoramic images and videos in Blender, YouTube. Available at: https://youtu.be/1qkldqKMvuM?si=1YtN3WCyJh6XzTKd. (Accessed: 04 February 2024).
Broken Frame Productions. (2019). Fulldome Tutorial: Convert 360o Video to Dome. YouTube. [online video] Available at: https://youtu.be/xM_PDDpaD_Y
Chill & Relax with Visual Effects (2021) VJ loops neon - abstract background video 4K - VJ loop 4K - colorful triangle background - HD, YouTube. Available at: https://youtu.be/wViaZrGMzoI?si=VjtvZKKjHdt2Jz_S (Accessed: 03 February 2024).
Epic Games (2019) Panoramic Capture Tool, in Unreal Engine | Unreal Engine 5.0 Documentation. Available at: https://docs.unrealengine.com/5.0/en-US/panoramic-capture-tool-in-unreal-engine/ (Accessed: 07 February 2024).
lumenandforge (2023) Everything you need to know about 360 domes, Lumen and Forge. Available at: https://lumenandforge.com/everything-you-need-to-know-about-360-domes/ (Accessed: 05 February 2024).
__________________________________________________________
Joe Abdel Sater
Polidomes.  (2022). Phillip Morris Projection Domes Presentation 2022. YouTube. [online video] Available at: https://www.youtube.com/watch?v=cYWJr1u8AU0
Fulldomepro. (2018). Inside the VR Dome. YouTube. [online video] Available at: https://youtu.be/5QOAILi0OZg
Broken Frame Productions. (2019). Fulldome Tutorial: Convert 360o Video to Dome. YouTube. [online video] Available at: https://youtu.be/xM_PDDpaD_Y
WINBUSH. (2022). Unreal Engine 5 360 Panoramic EASY! | No coding No Plugins. YouTube. [online video] Available at: https://youtu.be/SpkP1XCF_mE
Goddard, W., Muscat, A., Manning, J.G. and Jussi Holopainen (2016). Interactive dome experiences. doi: https://doi.org/10.1145/2994310.2994339  
____________________________________________________________
Anosh Ahmad
BBC (n.d.). BBC Sound Effects. [online] sound-effects.bbcrewind.co.uk. Available at: https://sound-effects.bbcrewind.co.uk/.
Running in the dark by Oliver Michael - Royalty Free music - artlist.io. Available at: https://artlist.io/royalty-free-music/song/running-in-the-dark/103684 (Accessed: 9th February 2024).
VideoCopilot.net After effects tutorials, plug-ins and stock footage for Post Production Professionals, Video Copilot. Available at: https://www.videocopilot.net/products/dsfx/ (Accessed: 12 February 2024)
Envato Elements. (n.d.). Neon Wormhole - hi-tech tunnel flythrough. [online] Available at: https://elements.envato.com/neon-wormhole-hi-tech-tunnel-flythrough-VZ2WEDN [Accessed 8th Feb. 2024].
‌____________________________________________________________
Abhinav
Ducky 3D 2019, Blender - easy sci-fi style animation loop in Eevee (Blender 2.8), YouTube, https://www.youtube.com/watch?v=mxUupHWoMLI.
Neb Motion 2020, Blender 2.82 - Hyper Space Tunnel 3D Animation Tutorial - Abstract Motion Graphic, YouTube, https://www.youtube.com/watch?v=-sU1WKRkpQ4.
0 notes
idp-dome · 3 months
Text
Bibliography
Talha Farooq
Blender Studio (2020) Geometry nodes from scratch, Blender Studio. Available at: https://studio.blender.org/training/geometry-nodes-from-scratch/ (Accessed: 05 February 2024).
Brandon’s Drawings (2022) Create 360 panoramic images and videos in Blender, YouTube. Available at: https://youtu.be/1qkldqKMvuM?si=1YtN3WCyJh6XzTKd. (Accessed: 04 February 2024).
Broken Frame Productions. (2019). Fulldome Tutorial: Convert 360o Video to Dome. YouTube. [online video] Available at: https://youtu.be/xM_PDDpaD_Y
Chill & Relax with Visual Effects (2021) VJ loops neon - abstract background video 4K - VJ loop 4K - colorful triangle background - HD, YouTube. Available at: https://youtu.be/wViaZrGMzoI?si=VjtvZKKjHdt2Jz_S (Accessed: 03 February 2024).
Epic Games (2019) Panoramic Capture Tool, in Unreal Engine | Unreal Engine 5.0 Documentation. Available at: https://docs.unrealengine.com/5.0/en-US/panoramic-capture-tool-in-unreal-engine/ (Accessed: 07 February 2024).
lumenandforge (2023) Everything you need to know about 360 domes, Lumen and Forge. Available at: https://lumenandforge.com/everything-you-need-to-know-about-360-domes/ (Accessed: 05 February 2024).
_______________________________________________________
Joe Abdel Satar
Polidomes.  (2022). Phillip Morris Projection Domes Presentation 2022. YouTube. [online video] Available at: https://www.youtube.com/watch?v=cYWJr1u8AU0
Fulldomepro. (2018). Inside the VR Dome. YouTube. [online video] Available at: https://youtu.be/5QOAILi0OZg
Broken Frame Productions. (2019). Fulldome Tutorial: Convert 360o Video to Dome. YouTube. [online video] Available at: https://youtu.be/xM_PDDpaD_Y
WINBUSH. (2022). Unreal Engine 5 360 Panoramic EASY! | No coding No Plugins. YouTube. [online video] Available at: https://youtu.be/SpkP1XCF_mE
Goddard, W., Muscat, A., Manning, J.G. and Jussi Holopainen (2016). Interactive dome experiences. doi: https://doi.org/10.1145/2994310.2994339  
________________________________________________________
Anosh Ahmad
BBC (n.d.). BBC Sound Effects. [online] sound-effects.bbcrewind.co.uk. Available at: https://sound-effects.bbcrewind.co.uk/.
www.videocopilot.net. (n.d.). VIDEO COPILOT | After Effects Tutorials, Plug-ins and Stock Footage for Post Production Professionals. [online] Available at: https://www.videocopilot.net/products/dsfx/ [Accessed 9 Feb. 2024].
Running in the dark by Oliver Michael - Royalty Free music - artlist.io. Available at: https://artlist.io/royalty-free-music/song/running-in-the-dark/103684 (Accessed: 9 February 2024).
Maxon. (n.d.). 3D Particle System Plugin for After Effects | Red Giant Trapcode…. [online] Available at: https://www.maxon.net/en/red-giant/trapcode/particular [Accessed 1 Feb. 2024].
‌_________________________________________________________
Abhinav
Ducky 3D 2019, Blender - easy sci-fi style animation loop in Eevee (Blender 2.8), YouTube, https://www.youtube.com/watch?v=mxUupHWoMLI.
Neb Motion 2020, Blender 2.82 - Hyper Space Tunnel 3D Animation Tutorial - Abstract Motion Graphic, YouTube, https://www.youtube.com/watch?v=-sU1WKRkpQ4.
_________________________________________________________
0 notes
idp-dome · 3 months
Text
Wrapping up Pre-Production
After attaining the information, from Sir.Ian Willcock regarding the actual shape and dimensions of the dome on 31st Jan, we had a couple of follow up meetings on 1st and 5th Feburary where we took the final roll call on all the creative and technical aspects governing the project. We had to give up on a couple of inital drafts we had in mind and take a new direction for the project. We made the following decisions :
Creative Outline: Creating a wormhole visualisation seemingly transporting our viewers across multiple dimensions filled with psychedelic visuals.
Technical Casting: a) Eric Bergner - Concept Design | Storyboarding b) Talha Farooq - Enviornment-1 | Base Project + Tecnhincal Lead c) Joe - Enviornment-2 | VFX + Light Design d) Abhinav - Enviornment-3 | Creature Animation e) Anosh - Enviornment 4 | Editing & Sound Design
Key Visual References: Thumper Game https://youtu.be/KqDqAIgYlrk?si=w-A2UgrdHIPADDJi Troubleshooting 101 :
Talha, Abhinav and Joe all highlighted some individual issues they were facing, regarding the rendering issues with the latest Unreal Engine, artefact distortion in the final output and transfering elements from Blender to Unreal etc. They guided one another to possible solutions or alternative avenues of troubleshooting and we worked towards having the entire on board Current Status:
Talha helped immensely by creating a base project in Unreal once again and guiding others on how to contribute and morph it for their own final outlook. Joe shared multiple tutorials and articles to help us all with understanding the technical requirements of the final output.
Eric has shared initial drafts of the storyboard and gave us a color pallete to play around with.
Im trying to create a narrative backbone for the video and trying to learn Cinema 4D to help garnish the final video with some VFX.
Abhinav is experimeting with creating creatures in Blender and incoporating them in the final project.
0 notes
idp-dome · 3 months
Text
Stage 1
For our following meetings, I began by presenting the research I had done over the weekend. I had come across a very useful article (https://doi.org/10.1145/2994310.2994339) that demonstrates one of the first dome interactive experiences that was catered for the game AstroSurf. Even though our project is not a game, the research process and its outcomes exposed me to the various technical aspects of creating a dome-based projection, which helped me start visualising the structure. Many concerns arose while reading the article, so I wanted to share them with my teammates. The first thing I noticed is that we did not know the definitive shape of the dome structure. Is it a semi-sphere resembling an igloo? Or is it a cylinder base with a flat dome ceiling? Is the video we create going to be projected from the top to the base of the walls, touching the floor? Or is it merely going to be projected on the top surface of the dome? All valid questions that we did not have answers to.
Another thing I presented to the group was the strong impact that the positioning of viewers has on the experience. In the research paper, there is a discussion about how people experiencing the simulation ended up lying on the floor instead of standing up and moving around. This exposed a greater need for designing the game in a way that makes it comfortable and easy for individuals to interact with. In our case, people will be lying on their backs looking up towards the top of the dome, but the problem is we don’t know in which direction their bodies will be positioned. For this reason, I suggested erasing the idea of having sky-ground, up-down, and ceiling-floor as recognizable perceptions. Instead, I proposed creating a spatial environment where the person is floating from one space to another. In this manner, the dome projection acts as a simulation rather than a mere representation of an image, a key concept when handling depiction and immersion. I also found a video which could help us learn how to convert panoramic renders from Unreal to a full-dome video in After Effects (https://www.youtube.com/watch?v=xM_PDDpaD_Y). It also gave the example of visualizing the projection in 3D using a virtual dome. Some videos I provided to better understand the technology at use were the like of https://www.youtube.com/watch?v=5QOAILi0OZg and https://www.youtube.com/watch?v=cYWJr1u8AU0.
On the other hand, Talha explained an experiment he had done regarding setting up an Unreal Engine scene and camera where he tested the rendered outcome using the Oculus VR headset. He proposed that motion sickness is one thing we need to be mindful of as camera movements can heavily impact the experience. He deduced that keeping the camera fixed, or only animating it on one axis would be preferable. He also provided us with another useful resource (https://lumenandforge.com/everything-you-need-to-know-about-360-domes/) which helped us understand the technology of 360 domes better. Talha also suggested the use of fade transitions in our video to go from one environment to another, but I felt like it would cause disruption to the experience, and make the simulation seem fragmented. Talha ended up setting up an Unreal Engine collaboration using GitHub source control and repositories, which we could use to work on the same project files from our own devices.
Anosh proposed using sound design elements to link the stages of our dome experience as a base for narration. He also set up the Tumblr blog for us and wrote an introductory post for it. He also explained how we could use it to submit posts to it from our own personal accounts. He also spent the time he had to get a bit familiar with Unreal Engine 5 and ask the rest of the team some questions in regards to the technicalities involved in our workflow.
Overall, we had agreed that we wanted to have colourful and vivid visuals that are eye-catching. Since we wanted to focus on the holistic experience, we thought of how we could use lighting to our advantage to direct the viewers’ attention and implement abstract geometric shapes in the landscape. For this reason, Eric developed some cool sketches and drawings that could inspire us to create 3D environments.
We had some alternate ideas that we wanted to keep as a plan B in case we decided to shift from our original concept:
Journey through desolate wastes (low-end soundscape, wind that ends up picking up as we ascend a mountain, Journey game)
Floating through space (slowly revealing some sort of hazard in space, maybe a wormhole)
Paralysed patient in lab or hospital (geometrical dimension)
Maybe combine all ideas through flashes. Basically Walking Simulator without the walking.
A jungle with mysterious creatures around.
To conclude, I ended up sharing a Pinterest board with pins I had gathered that could act as further inspiration or references to finalise our concept. (https://www.pinterest.com/itsjoeas/dome/?invite_code=0f2fdec8fce74f75aa46dbf6744733b7&sender=594756831948224619)
0 notes
idp-dome · 3 months
Text
Group One ... Assemble!
On January 24th, Sir Ian Willock, reached out to five key individuals from various Masters programs, in the University of Hertfordshire. He had a proposal for them: to band together and serve a greater purpose, by creating a 2 minute long, 3D Animated, interactive experience for a 180 degree dome, soon to hit their campus.
These 5 great men, battle hardened by last fall's semester and ready to sink their teeth in some assignments, jumped right in.
Later (the very same day), the world (or atleast their classmates) came to known as...Group One.
A team of highly skilled artists, in their own individual niche, combining their powers, time and energy to create a project, to grow their personal portfolio and more imporatantly, to entertain the masses en large.
So who are these extraordinary heroes (as no one likes to call them) you ask?
Eric Bergner Concept Artist, 2D Artist, experience in modeling with Maya as well as animation
Mohammad Talha Farooq Environment Art, 3D props- modeling in Blender, texture in substance, Unreal engine.
Abhinav 3D Modeling Auto Mobile experience with particles
Joe Abdel Sater Props - Environment Artist, Modeling in 3DS-Max and texturing in Substance.
Anosh Ahmad Video Editor, Sound Engineer, Animator, VFX Artist
What are their individual roles to play, how will they overcome the challenges ahead, will they achieve what destiny (or complete coincidence) brought them together for?
This blog, will cover their journey ahead.
1 note · View note