Research map / 3D vision / Depth and stereo
Revisiting 360 Depth Estimation With PanoGabor: A New Fusion Perspective
Zhijie Shen, Chunyu Lin, Lang Nie, Kang Liao, Weisi Lin, Yao Zhao
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 |
|---|---|---|---|
| add CS-UFM fusion module instead of concatenation (ablation) | PGFuse · Matterport3D | RMSE: 0.5021 -> 0.4906 | Improved |
| add PanoGabor layer to CS-UFM (ablation) | PGFuse · Matterport3D | RMSE: 0.4906 -> 0.4767 | Improved |
| add spherical gradient constraint (ablation) | PGFuse · Matterport3D | RMSE: 0.4767 -> 0.4624 | Improved |
| add spherical gradient constraint (ablation) | PGFuse · Matterport3D | delta1 (%): 88.51 -> 89.06 | Improved |
| use PanoGabor instead of normal Gabor filters (ablation) | PGFuse · Matterport3D | RMSE: 0.4848 -> 0.4624 | Improved |
| use normal Gabor filters in CS-UFM versus none (ablation) | PGFuse · Matterport3D | RMSE: 0.4906 -> 0.4848 | Improved |
| use distortion-aware Gabor initialization instead of Xavier (ablation) | PGFuse · Matterport3D | RMSE: 0.4856 -> 0.4624 | Improved |
| use distortion-aware initialization instead of handcrafted Gabor init (ablation) | PGFuse · Matterport3D | RMSE: 0.4848 -> 0.4624 | Improved |
| use random initialization of Gabor parameters (ablation) | PGFuse · Matterport3D | RMSE | Hurt |
| upgrade backbone from ResNet-18 to ResNet-34 (ablation) | PGFuse · Matterport3D | RMSE: 0.4933 -> 0.4624 | Improved |
| upgrade backbone from ResNet-34 to ResNet-50 (ablation) | PGFuse · Matterport3D | RMSE: 0.4624 -> 0.4618 | No clear effect |
| replace one cubemap with ERP features (ablation) | PGFuse · Matterport3D | RMSE: 0.4624 -> 0.4480 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Elite360D — Elite360D: Towards Efficient 360 Depth Estimation via Semantic- and Distance-Aware Bi-Projection Fusion (CVPR 2024) · better on 11 of 13 shared results
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
- BiFuse: Monocular 360 Depth Estimation via Bi-Projection Fusion (CVPR 2020 · similarity 0.87)
- Elite360D: Towards Efficient 360 Depth Estimation via Semantic- and Distance-Aware Bi-Projection Fusion (CVPR 2024 · similarity 0.85)
- RPG360: Robust 360 Depth Estimation with Perspective Foundation Models and Graph Optimization (NeurIPS 2025 · similarity 0.85)
- Splatter-360: Generalizable 360 Gaussian Splatting for Wide-baseline Panoramic Images (CVPR 2025 · similarity 0.85)
- Depth Anywhere: Enhancing 360 Monocular Depth Estimation via Perspective Distillation and Unlabeled Data Augmentation (NeurIPS 2024 · similarity 0.84)
- PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation (CVPR 2024 · similarity 0.84)
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.
