📝 Publications

SegmentDreamer: Towards High-fidelity Text-to-3D Synthesis with Segmented Consistency Trajectory Distillation
Jiahao Zhu, Zixuan Chen, Guangcong Wang, Xiaohua Xie, Yi Zhou.
IEEE/CVF International Conference on Computer Vision (ICCV), 2025.

SegmentDreamer is a novel framework designed to fully unleash the potential of consistency models for high-fidelity text-to-3D generation.

GuardSplat: Efficient and Robust Watermarking for 3D Gaussian Splatting
Zixuan Chen, Guangcong Wang, Jiahao Zhu, Jianhuang Lai, Xiaohua Xie.
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025.

GuardSplat aims to effectively protect the copyright of 3DGS assets, achieving superior capacity and fidelity to existing advances.

Hard-normal Example-aware Template Mutual Matching for Industrial Anomaly Detection
Zixuan Chen, Xiaohua Xie, Lingxiao Yang, Jianhuang Lai.
International Journal of Computer Vision (IJCV), 2024.

HETMM aims to address the hard mining issues in Industrial Anomaly Detection, significantly reducing the false alarms.

CuNeRF: Cube-Based Neural Radiance Field for Zero-Shot Medical Image Arbitrary-Scale Super Resolution
Zixuan Chen, Lingxiao Yang, Jianhuang Lai, Xiaohua Xie.
International Conference on Computer Vision (ICCV), 2023.

CuNeRF is the first zero-shot framework for Medical Image Arbitrary-Scale Super-Resolution.

Contour-aware loss: Boundary-aware learning for salient object segmentation
Zixuan Chen, Huajun Zhou, Jianhuang Lai, Lingxiao Yang, Xiaohua Xie.
IEEE Transactions on Image Processing (IEEE TIP), 2021.

This paper aims to employ contour cues and hierarchical attention modules to locate the salient objects.

Motion-appearance interactive encoding for object segmentation in unconstrained videos
Zixuan Chen, Chunchao Guo, Jianhuang Lai, Xiaohua Xie.
IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2020.

This paper aims to integrate motion and appearance cues to segment the moving object in video.

Interactive Two-Stream Decoder for Accurate and Fast Saliency Detection
Huajun Zhou, Xiaohua Xie, Jianhuang Lai, Zixuan Chen, Lingxiao Yang.
Computer Vision and Pattern Recognition Conference (CVPR), 2020.

This paper is a streamlined salient object detector that is beyond the real-time speed!

EvTrajGS: Accurate and Efficient 3D Gaussian Splatting from Unposed Event Streams
Zixuan Chen, Jiakai Zhang, Junhao Dong, Guangcong Wang, Jianhuang Lai, Yew-Soon Ong, Xiaohua Xie.

EvTrajGS enables accurate and efficient 3D reconstruction from unposed event streams through continuous-time camera trajectory modeling and joint pose-scene optimization.

VividDreamer: Towards High-Fidelity and Efficient Text-to-3D Generation
Zixuan Chen, Ruijie Su, Jiahao Zhu, Guangcong Wang, Lingxiao Yang, Jianhuang Lai, Shisong Wu, Xiaohua Xie.

VividDreamer aims to address the inherent trade-off between optimization costs and reconstruction quality, creating high-fidelity 3D assets in a short time.

APRF: Anti-Aliasing Projection Representation Field for Inverse Problem in Imaging
Zixuan Chen, Lingxiao Yang, Jianhuang Lai, Xiaohua Xie.

APRF aims to address the aliasing errors in Sparse-View Computed Tomography reconstruction.