UMAS3D: Unified Mask-Free 3D Editing with Region-Adaptive Source Preservation
cs.CV
Submitted: 2026-10-11
Updated: 2026-10-11
Project page: https://xianlin-zeng.github.io/UMAS3D
Sources
- EditP23: 3D Editing via Propagation of Image Prompts to Multi-View
- Native 3D Editing with Full Attention
- Generic 3D Diffusion Adapter Using Controlled Multi-View Editing
- MvDrag3D: Drag-based Creative 3D Editing via Multi-view Generation-Reconstruction Priors
- Easy3E: Feed-Forward 3D Asset Editing via Rectified Voxel Flow
- Velocity-Space 3D Asset Editing
- Feedforward 3D Editing via Text-Steerable Image-to-3D
- EditFlow3D: Automated Local Editing of 3D Assets with Trajectory Preservation
- DINOv2: Learning Robust Visual Features without Supervision
- GSEdit: Efficient Text-Guided Editing of 3D Objects via Gaussian Splatting
- Tailor3D: Customized 3D Assets Editing and Generation with Dual-Side Images
- DragGaussian: Enabling Drag-style Manipulation on 3D Gaussian Representation
- MIGE: Mutually Enhanced Multimodal Instruction-Based Image Generation and Editing
- Feedforward 3D Editing Learns from Semantic-Part Transformation
- Towards Scalable and Consistent 3D Editing
- Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models