Research map / 3D vision / Depth and stereo
Indoor Depth Recovery Based on Deep Unfolding with Non-Local Prior
Yuhui Dai, Junkang Zhang, Faming Fang, Guixu Zhang
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 the non-local auto-regressive (NLAR) module (ablation) | DC-NLAR · NYU-Depth V2 | RMSE: 0.108 -> 0.102 | Improved |
| add diagonal gradient loss Lgrad (ablation) | DC-NLAR · NYU-Depth V2 | RMSE: 0.103 -> 0.102 | Improved |
| use learned convolutional DC module instead of explicit 0/1 fidelity term (ablation) | DC-NLAR · NYU-Depth V2 | RMSE: 0.103 -> 0.102 | Improved |
| add NLAR, Lgrad and DC on top of the recovery module (ablation) | Recovery module baseline · NYU-Depth V2 | RMSE: 0.115 -> 0.102 | Improved |
| vary NLAR patch size (7, 9, 11) (ablation) | DC-NLAR · NYU-Depth V2 | RMSE: 0.102 / 0.102 / 0.103 | No clear effect |
| increase unrolling stages from 4 to 6 (ablation) | DC-NLAR · NYU-Depth V2 | RMSE: 0.102 -> 0.109 | Hurt |
| reduce unrolling stages from 4 to 2 (ablation) | DC-NLAR · NYU-Depth V2 | RMSE: 0.102 -> 0.107 | Hurt |
| DC-NLAR versus best prior method NLSPN | NYU-Depth V2 comparison · NYU-Depth V2 | RMSE: 0.092 -> 0.102 | Hurt |
Closest papers by meaning
- Non-Local Spatial Propagation Network for Depth Completion (ECCV 2020 · similarity 0.88)
- Guiding Monocular Depth Estimation Using Depth-Attention Volume (ECCV 2020 · similarity 0.85)
- Auto-Rectify Network for Unsupervised Indoor Depth Estimation (TPAMI 2021 · similarity 0.85)
- Efficient Indoor Depth Completion Network Using Mask-adaptive Gated Convolution (AAAI 2025 · similarity 0.85)
- Bilateral Propagation Network for Depth Completion (CVPR 2024 · similarity 0.84)
- Predicting High-precision Depth on Low-Precision Devices Using 2D Hilbert Curves (ICML 2025 · 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.
