ScenePlacer: Reformulating Scene-Level Object Pose Estimation as Visibility Field Learning
cs.CV
Submitted: 2026-10-11
Updated: 2026-10-11
Comments: 20 pages
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- HOLODECK 2.0: Vision-Language-Guided 3D World Generation with Editing
- Glob3R: Global Structure-from-Motion with 3D Foundation Models
- CAT3D: Create Anything in 3D with Multi-View Diffusion Models
- Hunyuan3D 2.1: From Images to High-Fidelity 3D Assets with Production-Ready PBR Material
- Any6D: Model-free 6D Pose Estimation of Novel Objects
- MV-SAM3D: Adaptive Multi-View Fusion for Layout-Aware 3D Generation
- CraftsMan3D: High-fidelity Mesh Generation with 3D Native Generation and Interactive Geometry Refiner
- Depth Anything 3: Recovering the Visual Space from Any Views
- Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks
- Lyra 2.0: Explorable Generative 3D Worlds
- ShapeR: Robust Conditional 3D Shape Generation from Casual Captures
- Stereo World Model: Camera-Guided Stereo Video Generation
- VGGT-$\Omega$
- World-R1: Reinforcing 3D Constraints for Text-to-Video Generation
- $\pi^3$: Permutation-Equivariant Visual Geometry Learning
- 3D-Fixer: Coarse-to-Fine In-place Completion for 3D Scenes from a Single Image
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