Haofei Xu

Papers

Total Citations

H-Index

About

Haofei Xu is a leading researcher in computer vision and graphics, with a focus on novel view synthesis, 3D scene representation, and immersive media. His most impactful work includes "PanSplat: 4K Panorama Synthesis with Feed-Forward Gaussian Splatting," a breakthrough method that enables high-resolution, real-time panorama generation from wide-baseline inputs. This work addresses critical challenges in virtual reality, autonomous driving, and robotics by combining the efficiency of 3D Gaussian splatting with feed-forward neural networks, achieving 4K resolution synthesis at unprecedented speeds. Xu’s contributions have garnered significant attention, with his papers accumulating over 3 citations in a short time, reflecting their immediate relevance to the field. His research bridges the gap between photorealistic rendering and practical deployment, offering scalable solutions for immersive applications. Notable achievements include advancing state-of-the-art in panorama synthesis, enabling fast inference without per-scene optimization, and setting new benchmarks for resolution and quality. Xu’s work continues to inspire innovations in spatial computing and real-time graphics, making him a rising star in the computer vision community.