Research map / 3D vision / Depth and stereo
DnD: Dense Depth Estimation in Crowded Dynamic Indoor Scenes
Dongki Jung, Jaehoon Choi, Yonghan Lee, Deokhwa Kim, Changick Kim, Dinesh Manocha, Donghwan Lee
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 flow-guided shape constraint and normal-guided scale constraint to the loss (ablation) | DnD · NAVERLABS Metro Station (F+B) | RMSE: 1.80 -> 1.76 | Improved |
| add flow-guided shape constraint and normal-guided scale constraint to the loss (ablation) | DnD · NAVERLABS Metro Station (human regions) | RMSE: 2.56 -> 2.44 | Improved |
| add normal-guided scale constraint (with FSC present) (ablation) | DnD · NAVERLABS Metro Station (human regions) | RMSE: 2.47 -> 2.44 | Improved |
| add flow-guided shape constraint (with NSC present) (ablation) | DnD · NAVERLABS Metro Station (human regions) | RMSE: 2.51 -> 2.44 | Improved |
| add flow-guided shape constraint (with NSC present) (ablation) | DnD · NAVERLABS Metro Station (F+B) | Abs Rel: 0.204 -> 0.189 | Improved |
| add normal-guided scale constraint (with FSC present) (ablation) | DnD · NAVERLABS Metro Station (F+B) | delta_1.25: 0.794 -> 0.806 | Improved |
| add depth loss, shape and scale constraints to photometric-only training (ablation) | DnD · NAVERLABS Metro Station (F+B) | RMSE: 2.39 -> 1.76 | Improved |
| add depth loss, shape and scale constraints to photometric-only training (ablation) | DnD · NAVERLABS Metro Station (human regions) | RMSE: 3.60 -> 2.44 | Improved |
| reduce the number of sparse input depth points (ablation) | DnD · NAVERLABS | delta_1.25 | Hurt |
| increase crowd density (12% or more human pixels) | DnD · NAVERLABS Metro Station and Department Store | delta_1.25: 6.2% and 6% | Hurt |
| increase crowd density (12% or more human pixels) | BTS · NAVERLABS Metro Station and Department Store | delta_1.25: 10.2% and 17.1% | Hurt |
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.
- Fast-MVSNet — Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement (CVPR 2020) · better on 3 of 3 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
- Estimating Depth from RGB and Sparse Sensing (ECCV 2018 · similarity 0.88)
- Dense Monocular Depth Estimation in Complex Dynamic Scenes (CVPR 2016 · similarity 0.88)
- R3D3: Dense 3D Reconstruction of Dynamic Scenes from Multiple Cameras (ICCV 2023 · similarity 0.87)
- Dense Dynamic Scene Reconstruction and Camera Pose Estimation from Multi-View Videos (ECCV 2026 · similarity 0.87)
- Depth on Demand: Streaming Dense Depth from a Low Frame Rate Active Sensor (ECCV 2024 · similarity 0.86)
- Seurat: From Moving Points to Depth (CVPR 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.
