3D Object Reconstruction Via Inverse Rendering
in Computer Graphics Last modified at:
This blog is coming soon…
Harnessing RAG in Game Development
in Machine Learning Last modified at:
Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for creating engaging narratives, dynamic dialogues, and immersive worlds in game development. Here’s a look at the technique and my hands-on experience implementing it during a summer at AgentLive.
As modern gaming demands more interactive storytelling and complex world-building, Retrieval-Augmented Generation (RAG) offers a potent solution. By integrating a retrieval mechanism with generative models, developers can craft dynamic narratives, non-player character (NPC) dialogues, and lore consistent with vast backstories and game design documents.
Reconstructing Century-Old Color Photos
in Computer Graphics Last modified at:
Sergei Mikhailovich Prokudin-Gorskii pioneered early color photography by capturing three-filtered monochrome images on glass plates. Modern image processing techniques can automatically restore these historical images with minimal artifacts.
Sergei Mikhailovich Prokudin-Gorskii (1863–1944) was a pioneering Russian photographer who foresaw the future of color photography. His method involved capturing three separate monochrome images using red, green, and blue filters on a single glass plate. When properly aligned and combined, these monochrome layers would produce a full-color image—an extraordinary feat at a time when color printing was challenging.
Procedural Generation of Fantasy Land
in Computer Graphics Last modified at:
Procedural generation offers a dynamic, efficient alternative to traditional modeling methods, creating diverse and intricate environments.
Procedural generation provides a solution to the challenges of traditional modeling, which can be time-consuming and lacks the variability needed for large-scale projects. This post explores a project developed by Andrew Yi and me for the Advanced Topics in Computer Graphics (Spring 2024) course at Yale University, where we aimed to create detailed, realistic 3D environments, focusing on urban landscapes and celestial bodies.
Can AI Detection Do Better
in Machine Learning Last modified at:
As large language models revolutionize natural language processing, there is an increasing need for robust detection methods to distinguish between AI-generated and human-written text.
The recent growth of large language models (LLMs) has led to an increase in the presence of AI-generated content on platforms. To maintain authenticity and prevent the spread of misinformation, it has become crucial to develop detection methods. DetectGPT and similar works have made progress in identifying machine-generated text using probability analysis. For the final project of CPSC 588 AI Foundation Models at Yale University, our group investigated and aimed to improve the existing model’s performance in detecting AI-generated sentences.
