EG360: Efficient and Generalizable 360 Depth Estimation via Adaptive Panorama-Perspective Distillation and Spherical Cross-Level Feature Fusion
cs.CV
Submitted: 2026-11-12
Updated: 2026-11-12
Code: https://github.com/facebookresearch/xformers
Terminology
Sources
- Joint 2D-3D-Semantic Data for Indoor Scene Understanding
- Unified Panoramic Geometry Estimation via Multi-View Foundation Models
- Matterport3D: Learning from RGB-D Data in Indoor Environments
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- Depth Anything in 360: Towards Scale Invariance in the Wild
- DA$^{2}$: Depth Anything in Any Direction
- Depth Anything 3: Recovering the Visual Space from Any Views
- Decoupled Weight Decay Regularization
- DINOv2: Learning Robust Visual Features without Supervision
- DINOv3
- DepthMaster: Unified Monocular Depth Estimation for Perspective and Panoramic Images
- DepthART: Scaling Foundation Monocular Depth to Tiny Models
- GoodSAM: Bridging Domain and Capacity Gaps via Segment Anything Model for Distortion-aware Panoramic Semantic Segmentation
- HoliCity: A City-Scale Data Platform for Learning Holistic 3D Structures
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs