Kerui Ren (original) (raw)

I am a second-year PhD candidate jointly supervised by Shanghai Jiao Tong University and Shanghai Artificial Intelligence Laboratory, under the primary guidance of Prof. Bo Dai. Currently, I am also a research intern at the Embodied AI Center of Shanghai AI Lab, co-advised by Dr. Mulin Yu and Dr. Linning Xu. Prior to my doctoral studies, I received my Bachelor of Engineering degree from Tongji University, where I conducted research on HDR reconstruction for dynamic scenes under the supervision of Prof. Zhangkai Ni.

My research interests lie in streaming reconstruction and scene generation. I have published papers in top-tier computer vision conferences and journals including CVPR, NIPS, ICLR, RSS, TPAMI, IJCV and TOG, with my publications accumulating 400+ Google Scholar citations.

🔥 News

• 2026.04: 🎉 One paper on Physically-based Light Transport was accepted to SIGGRAPH 2026.
• 2026.01: 🎉 Our ARTDECO on On-the-fly reconstruction was accepted to ICLR 2026.
• 2025.09: 🎉 Our MV-CoLight on Multi-view Object Compositing was accepted to NeurIPS 2025.
• 2025.08: 🎉 One paper on Unposed Novel View Synthesis was accepted to SIGGRAPH Asia 2025 (ACM TOG).
• 2025.06: 🎉 Our SMHDR on Multi-view HDR Reconstruction was accepted to IJCV 2025.
• 2025.05: 🎉 Our Octree-GS on LOD-Structured 3D Gaussians was accepted to TPAMI 2025.
• 2025.04: 🎉 One paper on One-Shot Manipulation was accepted to RSS 2025.
• 2025.02: 🎉 Our Horizon-GS on Aerial-to-Ground Scene Reconstruction was accepted to CVPR 2025.
• 2024.07: 🎓 Graduated from Tongji University.

📝 Publications

(†: corresponding author; * :equal contribution)

arxiv 2026

SoMA: A Real-to-Sim Neural Simulator for Robotic Soft-body Manipulation

Mu Huang, Hui Wang, Kerui Ren, Linning Xu, Yunsong Zhou, Mulin Yu, Bo Dai, Jiangmiao Pang

Project Code Paper

• SoMA is a Gaussian splat neural simulator that models deformable object dynamics from real-world robot manipulation, enabling action-conditioned, stable long-horizon simulation with high-fidelity, multi-view–consistent rendering.

arxiv 2026

PLANING: A Loosely Coupled Triangle-Gaussian Framework for Streaming 3D Reconstruction

Changjian Jiang*, Kerui Ren*, Xudong Li, Kaiwen Song, Guanghao Li, Linning Xu, Tao Lu, Junting Dong, Yu Zhang†, Bo Dai, Mulin Yu†

Project Code Paper

• PLANING introduces a loosely coupled triangle-Gaussian representation and a monocular streaming framework that jointly achieves accurate geometry, high-fidelity rendering, and efficient planar abstraction for embodied AI applications.

ICLR 2026

ARTDECO: Towards Efficient and High-Fidelity On-the-Fly 3D Reconstruction with Structured Scene Representation

Guanghao Li*, Kerui Ren*, Linning Xu, Zhewen Zheng, Changjian Jiang, Xin Gao, Bo Dai, Jian Pu†, Mulin Yu†, Jiangmiao Pang

Project Code Paper

• ARTDECO unifies 3D foundation priors with structured scene representations, enabling robust and generalizable 3D reconstruction of diverse real-world scenes using only monocular video.

SIGGRAPH Asia 2025 (ACM TOG)

AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Views

Lihan Jiang*, Yucheng Mao*, Linning Xu, Tao Lu, Kerui Ren, Yichen Jin, Xudong Xu, Mulin Yu, Jiangmiao Pang, Feng Zhao†, Dahua Lin, Bo Dai†

Project Code Paper

• We introduce AnySplat, a feed‑forward network for novel‑view synthesis from uncalibrated image collections in both sparse‑ and dense‑view scenarios.

CVPR 2025

Horizon-GS: Unified 3D Gaussian Splatting for Large-Scale Aerial-to-Ground Scenes

Lihan Jiang*, Kerui Ren*, Mulin Yu, Linning Xu, Junting Dong, Tao Lu, Feng Zhao, Dahua Lin, Bo Dai†

Project Code Paper

• We introduce Horizon-GS, tackles the unified reconstruction and rendering for aerial and street views with a new training strategy, overcoming viewpoint discrepancies to generate high-fidelity scenes.

arxiv 2025

Pay Attention to What You Need

Yifei Gao, Shaohong Chen, Lei Wang, Ruiting Dai, Ziyun Zhang, Kerui Ren, Jiaji Wu, Jun Cheng

Code Paper

• We propose Scaled ReAttention (SRA), a lightweight method that boosts LLMs’ long-context understanding without fine-tuning or retraining.

IJCV 2025

Semantic Masking with Curriculum Learning for Robust HDR Image Reconstruction

Zhangkai Ni, Yang Zhang, Kerui Ren, Wenhan Yang†, Hanli Wang†, Sam Kwong

Code

• We propose a novel SAM-guided MIM for HDR reconstruction (SMHDR). The combination of curriculum learning strategy and masked image modeling bridges the gap between pixel fitting and semantic understanding, which successfully turns the knowledge of SAM into the intrinsic understanding of the quality issues of HDR images.

📖 Educations

• 2024.09 - present, Ph.D. in Information and Communication Engineering, Shanghai Jiao Tong University
• 2020.09 - 2024.07, B.S. in Computer Science and Technology, Tongji University

Reviewer

• SIGGRAPH Asia 25, SIGGRAPH 26
• ISPRS, IEEE TVCG

💻 Internships

• 2023.11 - present, Shanghai Artificial Intelligence Laboratory, Shanghai, China.