Jiang Chaokang
World Models · Generative Driving · 3D/4D Perception
News
A lightweight timeline for papers, talks, releases, and long-term milestones.
I presented our Spotlight paper at COEX and wrote up five signals from the show floor — real-time streaming world models, one-step flow generation, closed-loop simulation, 3D geometry, and reliable multimodal evaluation.
Our VectorWorld work was specifically referenced and recognized in the invited keynote “Beneath the Cornerstone, Toward the Future: Automotive Intelligence and Global Expansion in the AI Era” delivered by Wu Yongqiao, President of Bosch Intelligent Driving Control (XC) China.
VectorWorld studies efficient streaming world models via diffusion flow on vector graphs.
RegFormer++ extends large-scale 3D LiDAR point registration with projection-aware 2D transformer design.
The journal version studies hierarchical diffusion models for uncertainty-aware 3D scene flow estimation.
The work explores odometry-assisted unsupervised 3D scene flow learning.
Mamba4D studies efficient long-sequence point cloud video understanding with disentangled spatial-temporal state space models.
The paper studies 4D neural fields and Gaussian deformation fields for point cloud interpolation.
A 4D point cloud interpolation framework based on neural fields and Gaussian deformation fields.
An efficient long-sequence point cloud video understanding model.
Invited by AutoDriver to share the CVPR 2024 work on 3D scene flow estimation and 3DSFLabelling.
The paper boosts 3D scene flow estimation by pseudo auto-labeling.
The paper studies robust uncertainty-aware scene flow estimation with iterative diffusion-based refinement.
Type to search across publications, projects, and blog notes.