Research map / 3D vision / Depth and stereo
DA$^{2}$: Depth Anything in Any Direction
Haodong Li, Wangguandong Zheng, Jing He, Yuhao LIU, Xin Lin, Xin Yang, Ying-Cong Chen, Chunchao Guo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| convert perspective depth datasets to panoramas and add to training (63K -> 607K samples) (ablation) | DA2 · Average of Stanford2D3D, Matterport3D, PanoSUNCG | AbsRel (%): 8.07 -> 6.62 | Improved |
| add Hypersim converted to panorama to Structured3D (ablation) | DA2 · Average of 3 benchmarks | AbsRel (%): 8.07 -> 7.10 | Improved |
| add Dynamic Replica panoramas (largest curated set) after 316,722 samples (ablation) | DA2 · Average of 3 benchmarks | AbsRel (%): 6.66 -> 6.62 | No clear effect |
| add panoramic out-painting (FLUX-I2P) of RGB images in curation (ablation) | DA2 · Average of 3 benchmarks | AbsRel (%): 7.59 -> 6.62 | Improved |
| add spherical embedding via cross-attention in SphereViT (ablation) | DA2 · Average of 3 benchmarks | AbsRel (%): 6.84 -> 6.62 | Improved |
| add auxiliary normal loss (ablation) | DA2 · Average of 3 benchmarks | AbsRel (%): 6.99 -> 6.62 | Improved |
| add spherical embedding via cross-attention in SphereViT (ablation) | DA2 · Average of 3 benchmarks | RMSE: 20.87 -> 20.63 | Improved |
| use P2E projection only for GT depth, no out-painting of depth | Data curation engine · Data curation design | depth GT accuracy | No clear effect |
| train end-to-end zero-shot DA2 vs strongest zero-shot baseline | DA2 · Zero-shot panoramic benchmarks | AbsRel: average 38% improvement | Improved |
| use L1 normal loss instead of angular loss | DA2 · Training loss design | training stability | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- UniK3D — UniK3D: Universal Camera Monocular 3D Estimation (CVPR 2025) · better on 6 of 6 shared results
- MoGev2 — MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision (CVPR 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Depth Any Panoramas: A Foundation Model for Panoramic Depth Estimation (CVPR 2026 · similarity 0.92)
- DepthMaster: Unified Monocular Depth Estimation for Perspective and Panoramic Images (NeurIPS 2026 · similarity 0.91)
- PanoVGGT: Feed-Forward 3D Reconstruction from Panoramic Imagery (CVPR 2026 · similarity 0.89)
- Splatter-360: Generalizable 360 Gaussian Splatting for Wide-baseline Panoramic Images (CVPR 2025 · similarity 0.89)
- PanDA: Towards Panoramic Depth Anything with Unlabeled Panoramas and Mobius Spatial Augmentation (CVPR 2025 · similarity 0.88)
- Unified Panoramic Geometry Estimation via Multi-View Foundation Models (NeurIPS 2026 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
