Relit-LiVE: Relight Video by Jointly Learning Environment Video

Weiqing Xiao*, Hong Li*, Xiuyu Yang*, Houyuan Chen, Wenyi Li, Tianqi Liu, Shaocong Xu, Chongjie Ye, Hao Zhao, Beibei Wang

[Paper] [Project] [Code]

• A video relighting framework that jointly predicts relit videos and viewpoint-aligned environment videos.
• Produces physically consistent and temporally stable relighting without requiring prior camera poses.

UniVidX: A Unified Multimodal Framework for Versatile Video Generation via Diffusion Priors

Houyuan Chen, Hong Li, Xianghao Kong, Tianrui Zhu, Shaocong Xu, Weiqing Xiao, Yuwei Guo, Chongjie Ye, Lvmin Zhang, Hao Zhao, Anyi Rao

[Paper] [Project]

• A unified multimodal video generation framework that leverages video diffusion priors for flexible conditional generation.
• Supports omni-directional generation across visual modalities with stochastic condition masking, decoupled gated LoRA, and cross-modal self-attention.

Simulate Any Radar: Attribute-Controllable Radar Simulation via Waveform Parameter Embedding

Weiqing Xiao*, Hao Huang*, Chonghao Zhong*, Yujie Lin, Nan Wang, Xiaoxue Chen, Zhaoxi Chen, Saining Zhang, Shuocheng Yang, Pierre Merriaux, Lei Lei, Hao Zhao

[Paper] [Project] [Code]

• A controllable radar cube simulation framework conditioned on customizable radar attributes.
• Supports radar attribute editing, actor removal, novel trajectories, and improves downstream perception tasks.

NeAR: Coupled Neural Asset-Renderer Stack

Hong Li*, Chongjie Ye*, Houyuan Chen, Weiqing Xiao, Ziyang Yan, Lixing Xiao, Zhaoxi Chen, Jianfeng Xiang, Shaocong Xu, Xuhui Liu, Yikai Wang, Baochang Zhang, Xiaoguang Han, Jiaolong Yang, Hao Zhao

[Paper] [Project]

• Couples a lighting-homogenized neural asset representation with a lighting-aware neural renderer.
• Enables single-image relightable 3D generation, novel-view relighting, and material-aware rendering.

Unifying Appearance Codes and Bilateral Grids for Driving Scene Gaussian Splatting

Nan Wang, Yuantao Chen, Lixing Xiao, Weiqing Xiao, Bohan Li, Zhaoxi Chen, Chongjie Ye, Shaocong Xu, Saining Zhang, Ziyang Yan, Pierre Merriaux, Lei Lei, Tianfan Xue, Hao Zhao

[Paper] [Project]

• Introduces multi-scale bilateral grids that unify appearance codes and bilateral grids.
• Improves geometric accuracy in dynamic driving-scene Gaussian Splatting by reducing photometric-inconsistency artifacts.

Uncertainty-Guided Iterative Architecture for Stereo Matching

Weiqing Xiao, Fengjun Zhong, Hao Zhao

[Paper]

• Introduces lightweight cost-volume-based disparity uncertainty estimation for iterative stereo matching.
• Uses uncertainty-guided update control and disparity rectification to improve stability, accuracy, and inference efficiency.

SR-Stereo & DAPE: Stepwise Regression and Pre-trained Edges for Practical Stereo Matching

Weiqing Xiao, Wei Zhao

[Paper] [Code]

• Introduces stepwise regression for robust stereo matching across domains with different disparity distributions.
• Proposes DAPE, a domain adaptation framework using pre-trained edge pseudo-labels to improve fine-tuning with sparse ground truth.