Islam Amar on LinkedIn: What Is Low-Rank Adaptation LoRA, or Low-Rank Adaptation, is a lightweight… (2024)

The Maze ProjectThe goal of this project was to construct a 3D game using ray-casting. I was involved in this project. This project's design and rendering work were completed over the period of three weeks. Ray-casting is a technique for converting a simple type of data into a 3D projection by tracing rays from the viewing volume to the view point.Why the maze Project?At some point, we've all loved playing realistic games on our phones and laptops. I know I have, with Temple Run 1 and 2 being my most recent passion (a little dated, I know). Ironically, I was never particularly interested in dabbling in game production. But when the opportunity to study and develop arose, I seized it. What a treat it would be to comprehend the inner workings of the games we like! I believe it has changed my perspective on these games.My Accomplishments :To create this maze, I used ray-casting to design the walls, leveraging LodeV's method of calculating ray length using vectors. Textures are used by default in the maze, although they can be disabled during execution. The maps are defined in text files as 2D arrays that are interpreted when supplied as an argument to the maze program. All other integers are shown as walls except 0 which depicts free space. In addition, we rendered and added textures using C and the SDL2 library.The following controls can be used to navigate the maze:⬆️- move forward⬅️- left➡️- right⬇️- move backward-ESC-exitWe implemented Wall Collision, which prevents camera rays from passing through the walls.Technical ChallengesLearning SDL2 was a difficult task for me that required a lot of attention on my part as I worked alone on it.Lesson learnt :For the first time in my life I came across terminology like ray-casting and SDL2 for the first time, but I was able to understand and work with them successfully. I wouldn't call myself a pro yet, but this was a significant step toward becoming a better game developer.If this endeavor taught me anything, it's that I'm willing to explore and journey into unfamiliar territory. To clarify, I never anticipated I'd bother learning anything about game production, but I jumped into this project head first and put my best foot forward. It certainly paid off. Following this, I am more willing to work on other games and master more advanced topics.About the authors :we are an ALX Software Engineering students, we are very passionate about coding and new technologies.Hajar AIT BAALIi: https://github.com/UHRAI/Rabab MCHICH: https://lnkd.in/eNW2bEsn project landing page : https://uhrai.github.io/project link : https://lnkd.in/egvjta94

Good morning! The link broke on my last post, so I'm reposting it here.I've designed a brand new Game Design Document, which can be used for free, for both personal and commercial purposes. It's a modular, comprehensive 32-page GDD. It is very modernized, as to meet the needs of game makers in 2023, including students and independent developers. Follow the link below to find the download. I hope this tool helps you to progress your career and aspirations in game production.Cheers!https://lnkd.in/gTue9ybG#gamedesign #gamedevelopment #3ddialectics

1

I was writing a script to document how I made my Bezel AR concept. I decided to get #chatGPT to change the tone to be similar to that of a Green Text story. Hilarity ensues:> be me, aspiring AR designer> wanna make AR thing everyone can use and share> not just for iPhone X gang, even potato phones welcome> use pixel art for 3D models> unleash cyberpunk and vaporwave vibes> make it use-anywhere friendly, not just couch-potato mode> start with moodboard.exe> sketch levels in Photoshop> 3D avatars in Vroids, cuz why not?> Blender for the heavy 3D stuff> put final touches in Figma, launch in Bezel> learningtime.jpg> feels good breaking all those pesky design rules> shoulda started in Bezel tho, 10/10 would recommend> Bezel > Reality Composer, Shapes XR, Unity> 6 hours in Bezel, but 12 days lost in 3D modeling abyss> not done yet, fam> future updates: back button and colliders for the winAnd that's the tale of my first AR concept. Thanks for listening to my Ted Talk.

10

5 Comments

Here's my tutorial for getting a gaussian splat bullettimeThis is something you can try at your next family gathering :)Sound on for instructions, see more tips below:Shooting TipsThe more people you have, the better this will turn out. Try to get an even coverage of different angles. It also works best if the phone models are really similar - (e.g. only a few years apart, or preferably the same year)Make sure everyone records slow motion, and records on 1x (don't use .5x or 2x lens options etc, unless *everyone* has the same option)Good lighting helps on a slow motion video. I shot facing a window that had good natural lightYou'll want to clap or have a visual cue, to make it easier to sync everything afterEditing TipsYou'll need to pull the same frame from each video, which is the most time consuming part. I did this in Premiere by lining up the clap/cue. You could also eyeball it, but the result will likely be more blurryI also did some edits and color correction to try to make the images more consistent. The more consistent your images look, the better the resultThen you can either edit the images together into a video and upload to the Luma AI app, or upload the images directly to the website (lumalabs.ai). It requires 10 images, but if you don't have 10 people, you can just take some of your images and duplicate them. The more images you have, the better the scan.Floaters cleanupI didn't clean up the floaters for my example, but if you prefer to you can use a desktop tool like Supersplat by PlayCanvas, the Luma Unreal plugin, or Unitydirector @karenxchengapp @luma_aiproducer @noellexms @thattiffwolffeaturing @priscillaylau @kevinzwhuang @mariajennings @ilnancy @svijnris

1,563

60 Comments

Metaverse(eISSN:2810-9791) is selected to be journal partner to support SIGGRAPH Asia 2024 Educator’s Forum.Submission Deadline: 7 August 2024, 23:59 AoE.Information details for who could submit or how to submit:https://lnkd.in/eywGE9kHSIGGRAPH Asia 2024 Educator’s Forum invites researchers, developers and educators from both academia and the industry to present innovative research outcomes, methods and developments related to the educations of computer graphics and interactive techniques and application of computer graphics and interactive techniques in all areas of learning.The program is seeking submissions for:.Education Papers: Peer-reviewed academic papers that present research and case studies relevant to the SIGGRAPH Asia Educator’s Forum..Education Talks: Short presentations covering newly developed works, ideas, and projects still in the development stage..EETA(Engaging Education Techniques and Assignments): Short presentations of engaging class assignments on computer graphics and interactive techniquesEducation Papers and Education Talksshould build upon previous research, reference related works, and include an evidence-based discussion on the educational benefits. Best education paper awards will be selected by the committees among the presented education papers and talks in SIGGRAPH Asia 2024.We encourage submissions in educational research in the scope of SIGGRAPH Asia conference, including Computer Graphics, Interactive Techniques, CG Animation, Video Games, VFX, VR/AR/XR, Visualization, AI in Graphics:.Application of these areas to education of other topics, such as Generative AI, STEAM, fine arts, performing arts, design, architecture, medical and social sciences.Curriculum development and the use of learning techniques for teaching these areas.Teaching of these areas

The purpose of doing this project was to create a game in 3D using ray-casting. I worked on this project. The tasks of this project were done over the course of 3 weeks as pertaining to designing and rendering. Ray-casting is technique that transform a simple form of data into a 3D projection by tracing rays from the view point into the viewing volume.Why the Maze Project?We’ve all enjoyed playing realistic games on our phones and computers at some point. I know I have, with my most recent obsession being Temple Run 1 and 2 (a little outdated, I know). Ironically, I never was really interested in trying my hand at game development. However, when the opportunity to learn and build presented itself I took it. How nice it would be to understand the back-end of the games we play! I think it makes me view these games in a whole different light.Architecture DiagramTo make this maze, I used ray-casting to draw the walls, utilizing LodeV’s method of using vectors to calculate ray length. By default the maze uses textures but textures can be disabled on execution. The maps are defined in 2D arrays in text files, which are parsed when passed as an argument to the maze executable. 0 represents open space, all other integers are drawn as walls. Additionally, we used C language and the SDL2 library to render and add the textures.To navigate through the maze, the following controls can be used:· W : move forward· S : move backward· A : rotate camera left· D : rotate camera right· Q : strafe left· E : strafe right· F : toggle full-screen· ESC : quit mazeWall Collision is a feature we implemented which doesn’t allow camera rays to pass through the walls.Technical ChallengesLearning SDL2 was quite a challenge for me and required a lot of dedication from me to understand. Working together with my teammates and discussing helped me massively.Also ensuring we have installed the correct and necessary SDL2 packages to make the window was a difficulty my entire team faced and had a challenge trying to fix.Lessons LearntThe greatest lesson I learnt from this project is something I already mentioned in the beginning of this blog post: I was able to develop a game.For the first time I came across terms like ray-casting and SDL2, yet I was able to understand them and work with them successfully. I wouldn’t say I am a pro yet but this was surely a stepping stone to becoming better at game dev.If there’s anything this project taught me about myself, it’s that I am willing to explore and venture into unknown territories. To reiterate, I never imagined I would bother to know anything with regards to game development, but I took on this project head on without hesitation and put my best foot forward. It definitely paid off. I am more open to working on more games and learning deeper concepts after this.github link: https://lnkd.in/ddmgEZNp

this is the right path of future thinking

In the pursuit of my diplomawork 🎓, I embarked on an ambitious project 🚀 to create a videogame prototype 🎮, diving deep into the intricacies of 3D modeling 🖥️ and texturing 🎨. This endeavor was not just about learning the techniques; it was about mastering them to their core 🌟 and developing a nuanced methodology for their application 🛠️.A revelation of my project was the impact of reference 📚 on the quality of 3D models. This was an exploration into how to seek 🧐 and work from reference materials. This insight became a cornerstone of my creative process both in 2D drawing ✏️ and 3D.My diploma work aimed to capture the feeling of a free-thinking mind 🧠, uninfluenced by faulty logic 🤔, lies 🤥, mischief 😈, or propaganda 📣. It’s like the open-mindedness of childhood 👶, a feeling I’ve noticed in my life but couldn’t always define.In my game, every character’s design reflects its concept 🎭. Instead of using story 📖 and cutscenes 🎬, the game conveys its theme through environment 🌍, characters 👥, and animation 🎞️. This approach inspired the creation of large, slow mechanical robots 🤖 and quick, plasma-like entities ⚡, offering visual expressiveness 🌈.The prototype was a dynamic platform 🕹️. Utilizing Unity 🌐, I frequently made real-time in-game presentations 🎥 to exhibit the game’s development 📈.In the development of my videogame prototype, I was confronted with a challenging aspect of character design: animating the character’s unique plasma limbs ⚡ so they could interact with the environment in real time. I couldn’t just make something in maya and then import it into unity. Innovation was needed 💡.This led me to discover an online spline tool 🔍 and to delve into opengl/directx shader programming 🖱️. Integrating these tools into Unity, I was able to experiment with new ways of animating the character’s limbs 🎨, aiming for a more fluid and interaction with the game’s environment. Although it stayed on the prototype level, it’s a deeply interesting topic and work 🌟. On top of all of this this project gave me the chance to consult and cooparate with Pápai Bence. Who in my opinion makes the most fun indiegames in hungary.MORE ON MY INSTAGRAM: https://lnkd.in/dgND6wsnYOUTUBE: https://lnkd.in/d9v94QSS

4

Apple's innovation engine is at it again with "Keyframer," a groundbreaking tool that's changing the game for animators and creatives alike. Imagine turning static images into dynamic animations, all through the power of language prompts. With Keyframer, users breathe life into Scalable Vector Graphics (SVGs) using just their words. Describe what you want to see in motion, and voilà — your vision becomes animated reality. It's like having a conversation with your art. Thanks to the genius of GPT-4, Keyframer translates your natural language prompts into CSS animations. This means you can literally "talk" your ideas into existence, refining and building complex animations with ease. Expert or newbie, Keyframer speaks your language. Keyframer isn't just a tool; it's a revolution in animation and creative expression. By democratizing animation creation, it lowers the barriers to entry and opens up a world of possibilities for anyone with a vision. The future of creativity is here, and it's accessible.

1

Want to turn your next family gathering into a playful experience? 🎬 Check how Karen X. Cheng unwraps a simple technique for creating a 3D family photo, reminiscent of that Matrix moment when Neo dodges bullets. #3DGaussianThe best part is that instead of professional cameras, you just need your family or friends with their smartphones. Utilizing the 3D Gaussian splatting technique, these smartphone-captured images undergo processing to produce fun 3D photos using Luma AI.Feel free to connect with Reflektor Digital if you want to learn more about 3D Gaussian splatting!

7