Scholar
Pengzhi Li
Google Scholar ID: ttBQptwAAAAJ
Tsinghua University
Citations & Impact
Citations
253
H-index
8
i10-index
8
Publications
13
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Publications: ART3D: 3D Gaussian Splatting for Text-Guided Artistic Scenes Generation (CVPR 2024); Tuning-Free Image Customization with Image and Text Guidance (ECCV 2024); LayerDiffusion: Layered Controlled Image Editing with Diffusion Models (SIGGRAPH Asia 2023, ICML 2023); Sketch-to-Architecture: Generative AI-aided Architectural Design (Pacific Graphics 2023); Generating Daylight-driven Architectural Design from Massing Models via Diffusion Models; Towards Practical Consistent Video Depth Estimation (ICMR 2023); Efficient Temporal Denoising for Improved Depth Map Applications (ICLR 2023).
Research Experience
- Tencent Youtu Lab (Shenzhen): Research Intern, Aug. 2023 - Jan. 2024, Topic: Image Generation and Editing; University of Pennsylvania (Remote): Research Intern, Apr. 2023 - Jun. 2023, Topic: Machine Learning and Statistics; Huawei 2012 Lab (China): Research Intern at Media Technology Institute, Jun. 2022 - Dec. 2022, Topic: Video Generation and Depth Estimation.
Education
- Master's degree: Electronic Information and Engineering at Tsinghua University, Sep. 2021 - Jun. 2024, Advisor: Prof. Zhiheng Li; Bachelor's degree: Architecture from CQJTU
Background
- Research interests: Intersection of computer science and arts & design, focusing on generative models and their applications in image/video/3D generation and graphic design. Earlier research also involved 3D vision.
Miscellany
- Excellence in painting, 3D modeling, and graphic design, with awards in several design competitions. Amateur photographer specializing in landscape photography. Passionate about painting, 3D software modeling (e.g., Maya, Blender, UE, Rhino), computer graphics, graphic design, and architectural design.
Co-authors