3D Generation Comparison Hub
4 papers - avg viability 6.5
Top Papers
- MV-SAM3D: Adaptive Multi-View Fusion for Layout-Aware 3D Generation(9.0)
MV-SAM3D enhances 3D generation by integrating multi-view consistency and physical plausibility without additional training.
- RelaxFlow: Text-Driven Amodal 3D Generation(7.0)
RelaxFlow is a training-free framework for text-driven amodal 3D generation that uses a dual-branch approach to decouple control granularity, enabling completion of unseen regions based on text prompts while preserving input observations.
- Cog2Gen3D: Sculpturing 3D Semantic-Geometric Cognition for 3D Generation(7.0)
Cog2Gen3D is a 3D cognition-guided diffusion framework for controllable and physically plausible 3D generation, leveraging semantic and geometric information.
- Hoi3DGen: Generating High-Quality Human-Object-Interactions in 3D(3.0)
Hoi3DGen generates high-quality 3D human-object interactions from text for AR, XR, and gaming applications.