Multimodal LLMs · Video World Models · 3D Spatial Understanding多模态大模型 · 视频世界模型 · 三维空间理解
likai211@mails.ucas.ac.cn·Google Scholar·GitHub·ORCID·LinkedIn·CV in English (PDF)英文简历 (PDF)·CV in Chinese (PDF)中文简历 (PDF)
My research centres on large multimodal models and how they acquire spatial and temporal understanding. I am currently working on memory in video world models, alongside reinforcement learning for sparse-view 3D scene reconstruction. Earlier, during my PhD, I proposed the Offset Token, which brought off-nadir photogrammetry into the perceptual scope of vision foundation models.我的研究以多模态大模型为核心,关注模型如何获得空间与时间层面的理解能力。目前正在研究视频世界模型(Video World Model)中的记忆机制,并同时探索用强化学习实现稀疏视角三维场景重建。博士期间,我提出了偏移量词元(Offset Token)的概念,将偏移摄影带入视觉大模型的理解视角。
[1] Prompt-Driven Building Footprint Extraction in Aerial Images with Offset-Building Model
[2] PolyFootNet: Extracting Polygonal Building Footprints in Off-Nadir Remote Sensing Images
[3] DragOSM: Extract Building Roofs and Footprints from Aerial Images by Aligning Historical Labels
[4] ObliCity: A Benchmark and Baseline for Roof-to-Ground Projection Displacement Correction
[6] Scenix: Sparse-View 3D Scene Reconstruction via Executable Scene Programs
· Reinforcement-Learning-Driven, Sparse-View-Consistent Indoor Scene Reconstruction基于强化学习的稀疏视角一致性室内场景重建
· Memory Mechanisms for Video World Models视频世界模型的记忆机制
[7] Remote-Sensing City Layout Extraction with MLLM
[10] IRSAMap: Towards Large-Scale, High-Resolution Land Cover Map Vectorization
[12] LRSCLIP: A Vision-Language Foundation Model for Aligning Remote Sensing Image with Longer Text
[13] SayAnything: Audio-Driven Lip Synchronization with Conditional Video Diffusion
[14] CusMer: Multimodal Intent Recognition in Customer Service via Data Augment and LLM Merge
[15] SAMPolyBuild: Adapting the Segment Anything Model for polygonal building extraction
[16] 基于边界曲线的多因素机场出租车管理系统模型
[17] Land Price Assessment Based on Deep Neural Network
[18] Urban Functional Regions Discovering Based on Deep Learning
[19] A WordNet-Based Geospatial Web Services Search Method Supporting Quality of Service Constraints
UCAS · AIRCAS · CityU AML Lab中国科学院大学 · 空天信息创新研究院 · 香港城市大学 AML 实验室
Beijing, China / Hong Kong SAR中国北京 / 中国香港
likai211@mails.ucas.ac.cn
WeChat: kaili37