5 papers
Variation-aware Flexible 3D Gaussian Editing
Hao Qin, Yukai Sun, Meng Wang +3
Indirect editing methods for 3D Gaussian Splatting (3DGS) have recently witnessed significant advancements. These approaches operate by first applying edits in the rendered 2D spac…
Linking Perception, Confidence and Accuracy in MLLMs
Yuetian Du, Yucheng Wang, Rongyu Zhang +5
Recent advances in Multi-modal Large Language Models (MLLMs) have predominantly focused on enhancing visual perception to improve accuracy. However, a critical question remains une…
Style-Aligned Image Composition for Robust Detection of Abnormal Cells in Cytopathology
Qiuyi Qi, Xin Li, Ming Kong +4
Challenges such as the lack of high-quality annotations, long-tailed data distributions, and inconsistent staining styles pose significant obstacles to training neural networks to…
Distilling Multi-view Diffusion Models into 3D Generators
Hao Qin, Luyuan Chen, Ming Kong +2
We introduce DD3G, a formulation that Distills a multi-view Diffusion model (MV-DM) into a 3D Generator using gaussian splatting. DD3G compresses and integrates extensive visual an…
Probablistic Restoration with Adaptive Noise Sampling for 3D Human Pose Estimation
Xianzhou Zeng, Hao Qin, Ming Kong +2
The accuracy and robustness of 3D human pose estimation (HPE) are limited by 2D pose detection errors and 2D to 3D ill-posed challenges, which have drawn great attention to Multi-H…