Haodong Li (李浩东)

I'm currently a CS PhD student at UC San Diego, working with Prof. Manmohan Chandraker. Prior to this, I got my MPhil degree in CS from HKUST, working with Prof. Ying-Cong Chen, and my BEng degree in Automation from Zhejiang University. I spent a wonderful summer at Tencent as a student researcher.

I'm currently exploring topics on 3D / video / agents (e.g., LLM, VLM, VLA) tasks. I'm very open to collaborations. Just drop me an email if you're interested! 😊

Email: hal211@ucsd.edu
Wechat: lihaodong-2000

Google Scholar  /  Github  /  Twitter (X)  /  Linkedin

Selected Publications

arXiv 2025
arXiv (Soon) / Paper / Project Page / Github (Soon) / Data (Soon) / Demo (Soon) / Slides

Powered by large-scale training data curated from our panoramic data curation engine, and the SphereViT for addressing the spherical distortions in panoramas, DA² is able to predict dense, scale-invariant distance from a single 360° panorama in an end-to-end manner, with remarkable geometric fidelity and strong zero-shot generalization.

ICLR 2025
arXiv / Paper / Project Page / Github / Demo (D) / Demo (N) / ComfyUI

Lotus is a diffusion-based visual foundation model with a simple yet effective adaptation protocol, aiming to fully leverage the pre-trained diffusion's powerful visual priors for dense prediction. With minimal training data, Lotus achieves SoTA performance in two key geometry perception tasks, i.e., zero-shot monocular depth and normal estimation.

ICLR 2025
arXiv / Paper / Project Page / Github / Demo

Characterized by its emphasis on the interpretation of subject-essential attributes, the proposed DisEnvisioner effectively identifies and enhances the subject-essential feature while filtering out other irrelevant information, enabling exceptional image customization without cumbersome tuning or relying on multiple reference images.

SIGGRAPH Asia 2024
Paper / Video

DIScene is capable of generating complex 3D scene with decoupled objects and clear interactions. Leveraging a learnable Scene Graph and Hybrid Mesh-Gaussian representation, we get 3D scenes with superior quality. DIScene can also flexibly edit the 3D scene by changing interactive objects or their attributes, benefiting diverse applications.

ECCV 2024
arXiv / Paper / Project Page

We introduce Bi-TTA, a method that leverages spatial and temporal consistency for appropriate self-supervision, coupled with novel prospective and retrospective adaptation strategies, enabling superior adaptation ability of pre-trained rPPG models to the target domain using only unannotated, instance-level target data.

CVPR 2024 Highlight
arXiv / Paper / Github / Demo / Video

We present LucidDreamer, a text-to-3D generation framework, to distill high-fidelity textures and shapes from pretrained 2D diffusion models with a novel Interval Score Matching objective and an advanced 3D distillation pipeline. Together, we achieve superior 3D generation results with photorealistic quality in a short training time.

Education

Experience

Thanks for visiting!