Research map / 3D vision / Depth and stereo
Toward Practical Monocular Indoor Depth Estimation
Cho-Ying Wu, Jialiang Wang, Michael Hall, Ulrich Neumann, Shuochen Su
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 |
|---|---|---|---|
| add DPT depth-structure distillation to left-right self-supervised training (ablation) | DistDepth on VA · VA | MAE: 0.295 -> 0.253 | Improved |
| use the larger DPT-legacy expert rather than DPT-hybrid (ablation) | DistDepth on VA · VA | MAE: 0.276 -> 0.253 | Improved |
| add test-time multi-frame cost-volume inference | DistDepth on VA · VA | MAE: 0.253 -> 0.239 | Improved |
| add statistical depth-structure distillation (ablation) | DistDepth · qualitative depth-structure study | depth structure | Improved |
| add spatial boundary refinement to statistical distillation (ablation) | DistDepth · qualitative depth-structure study | depth structure | Improved |
| use a low turn-on gradient quantile instead of a high one for spatial refinement (ablation) | DistDepth · qualitative depth-structure study | boundary structure | Hurt |
| train DistDepth on SimSIN rather than UniSIN, without NYUv2 training | DistDepth on NYUv2 · NYUv2 test | AbsRel: 0.158 -> 0.164 | Hurt |
| train on SimSIN instead of UniSIN | DistDepth on real data · real indoor evaluation | geometric depth quality | No clear effect |
| finetune DistDepth on NYUv2 using temporal consistency | DistDepth on NYUv2 · NYUv2 test | AbsRel: 0.164 -> 0.130 | Improved |
Closest papers by meaning
- StructDepth: Leveraging the Structural Regularities for Self-Supervised Indoor Depth Estimation (ICCV 2021 · similarity 0.90)
- Geometric Structure Based and Regularized Depth Estimation From 360 Indoor Imagery (CVPR 2020 · similarity 0.89)
- Guiding Monocular Depth Estimation Using Depth-Attention Volume (ECCV 2020 · similarity 0.89)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.88)
- RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptions (NeurIPS 2024 · similarity 0.87)
- A Simple yet Universal Framework for Depth Completion (NeurIPS 2024 · similarity 0.87)
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.
