Xiaoyu Xiang

Papers

Total Citations

H-Index

About

Xiaoyu Xiang is a leading researcher in computer vision and generative AI, with a focus on photorealistic 3D content creation. Their most notable contribution is the development of **CAD (Consistent Adversarial Distillation)**, a groundbreaking framework for photorealistic 3D generation. Unlike traditional Score Distillation Sampling (SDS) methods, which often produce blurry or inconsistent results, CAD leverages adversarial distillation to enforce high-fidelity, multi-view consistency, enabling the synthesis of detailed 3D objects suitable for AR/VR, robotics, and gaming. This work, published in 2024, has already garnered 4 citations, signaling its rapid impact on the field. Xiang’s research addresses the critical challenge of bridging 2D generative models with 3D representations, pushing the boundaries of what is possible in automated 3D asset creation. By tackling the limitations of prior diffusion-based pipelines, their work promises to democratize 3D content generation, making it more accessible for real-world applications. Xiang continues to advance the frontier of generative modeling, with a clear trajectory toward shaping the future of immersive digital experiences.