Research map / 3D vision / Depth and stereo
BiFuse: Monocular 360 Depth Estimation via Bi-Projection Fusion
Fu-En Wang, Yu-Hsuan Yeh, Min Sun, Wei-Chen Chiu, Yi-Hsuan Tsai
Check 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 |
|---|---|---|---|
| Use the two-branch BiFuse model with learnable bi-projection fusion | BiFuse · Matterport3D | MAE: 0.4008 -> 0.3470 | Improved |
| Use the two-branch BiFuse model with learnable bi-projection fusion | BiFuse · Stanford2D3D | MAE: 0.3428 -> 0.2343 | Improved |
| Use the two-branch BiFuse model with learnable bi-projection fusion | BiFuse · PanoSUNCG | MAE: 0.1346 -> 0.0789 | Improved |
| Use the two-branch BiFuse model with learnable bi-projection fusion | BiFuse · 360D | MAE: 0.1381 -> 0.1143 | Improved |
| Add spherical padding rather than zero padding or cube padding (ablation) | Cubemap branch · Matterport3D | RMSE: 0.6934 -> 0.6327; 0.6628 -> 0.6327 | Improved |
| Add spherical padding rather than zero padding or cube padding (ablation) | Cubemap branch · Stanford2D3D | RMSE: 0.4511 -> 0.4269; 0.4407 -> 0.4269 | Improved |
| Add spherical padding rather than zero padding or cube padding (ablation) | Cubemap branch · PanoSUNCG | RMSE: 0.3441 -> 0.2874; 0.2946 -> 0.2874 | Improved |
| Add spherical padding rather than zero padding or cube padding (ablation) | Cubemap branch · 360D | RMSE: 0.2819 -> 0.2614; 0.2722 -> 0.2614 | Improved |
| Use learnable masked fusion instead of direct feature addition (ablation) | BiFuse on Matterport3D · Matterport3D | MAE: 0.4842 -> 0.3470 | Improved |
| Use learnable masked fusion instead of feature averaging (ablation) | BiFuse on Matterport3D · Matterport3D | MAE: 0.4405 -> 0.3470 | Improved |
Closest papers by meaning
- BiFuse++: Self-Supervised and Efficient Bi-Projection Fusion for $360^{\circ }$ Depth Estimation (TPAMI 2022 · similarity 0.91)
- Elite360D: Towards Efficient 360 Depth Estimation via Semantic- and Distance-Aware Bi-Projection Fusion (CVPR 2024 · similarity 0.88)
- Improving 360 Monocular Depth Estimation via Non-local Dense Prediction Transformer and Joint Supervised and Self-Supervised Learning (AAAI 2022 · similarity 0.88)
- Depth Anywhere: Enhancing 360 Monocular Depth Estimation via Perspective Distillation and Unlabeled Data Augmentation (NeurIPS 2024 · similarity 0.87)
- RPG360: Robust 360 Depth Estimation with Perspective Foundation Models and Graph Optimization (NeurIPS 2025 · similarity 0.86)
- Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025 · similarity 0.85)
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.
