Research map / 3D vision / Depth and stereo
DELTAR: Depth Estimation from a Light-Weight ToF Sensor and RGB Image
Yijin Li, Xinyang Liu, Wenqi Dong, Han Zhou, Hujun Bao, Guofeng Zhang, Yinda Zhang, Zhaopeng Cui
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 sampled depth hypotheses + PointNet instead of mean-variance features (ablation) | DELTAR · ZJU-L5 test | delta1: 0.434 -> 0.853 | Improved |
| use sampled depth hypotheses instead of five-channel input (RGB, mean, variance) (ablation) | DELTAR · ZJU-L5 test | delta1: 0.619 -> 0.853 | Improved |
| add transformer fusion instead of feature concatenation (ablation) | DELTAR · ZJU-L5 test | delta1: 0.825 -> 0.853 | Improved |
| restrict cross-attention to corresponding patch-distribution pairs (ablation) | DELTAR · ZJU-L5 test | delta1: 0.749 -> 0.853 | Improved |
| add image self-attention (ablation) | DELTAR · ZJU-L5 test | delta1: 0.835 -> 0.853 | Improved |
| add image-to-distribution attention (ablation) | DELTAR · ZJU-L5 test | delta1: 0.840 -> 0.853 | Improved |
| sample depth hypotheses by inverse CDF instead of uniformly (ablation) | DELTAR · ZJU-L5 test | RMSE: 0.439 -> 0.436 | Improved |
| add mViT refinement module (ablation) | DELTAR · ZJU-L5 test | RMSE: 0.462 -> 0.436 | Improved |
| add mViT refinement module (ablation) | DELTAR · ZJU-L5 test | delta1: 0.850 -> 0.853 | No clear effect |
Closest papers by meaning
- FloatingFusion: Depth from ToF and Image-Stabilized Stereo Cameras (ECCV 2022 · similarity 0.88)
- Multi-Modal Neural Radiance Field for Monocular Dense SLAM with a Light-Weight ToF Sensor (ICCV 2023 · similarity 0.88)
- RGB Guided ToF Imaging System: A Survey of Deep Learning-Based Methods (IJCV 2024 · similarity 0.87)
- Dense Metric Depth Completion from Sparse Direct Time-of-Flight Sensors (CVPR 2026 · similarity 0.87)
- DEPTHOR: Depth Enhancement from a Practical Light-Weight dToF Sensor and RGB Image (ICCV 2025 · similarity 0.86)
- Cross-spectral Gated-RGB Stereo Depth Estimation (CVPR 2024 · similarity 0.83)
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.
