Research map / 3D vision / Depth and stereo
LiteSense: Lifting Lightweight ToF with RGB for High-Resolution Metric Depth Estimation
Yusheng Li, Lizhi Lou, Yan Tang, Zekai Miao, Shaoming Zhang, Jianmei Wang
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 ToF depth and CNH histogram with PCSI fusion to an RGB-only network (ablation) | lightweight RGB baseline · NYUv2 | delta1 / AbsRel: 0.663 -> 0.982; 0.216 -> 0.029 | Improved |
| use native CNH distribution instead of depth-only or DELTAR sampled distribution (ablation) | LiteSense auxiliary input · NYUv2 | accuracy | Improved |
| concatenate upsampled ToF depth to RGB as an input channel (ablation) | LiteSense · NYUv2 | accuracy | Improved |
| replace patch-wise PCSI with global cross-attention between RGB-D and CNH features (ablation) | LiteSense · NYUv2 | accuracy | Hurt |
| fuse CNH only at 1/2 and 1/4 decoder scales rather than all stages | LiteSense decoder · NYUv2 | artifacts | Improved |
| use LiteSense instead of DELTAR for RGB-ToF fusion | DELTAR · NYUv2 | delta1 / AbsRel: delta1 0.952 -> 0.982; AbsRel 0.064 -> 0.029 | Improved |
| use LiteSense instead of DELTAR | DELTAR · NYUv2 | parameters: 18.55 -> 5.48 M | Improved |
| add ToF auxiliary priors to metric depth estimation (zero-shot) | monocular baselines · SUN RGB-D zero-shot | delta1: Metric3D-V2-L 0.962 -> LiteSense 0.980 | Improved |
| fine-tune on real sensor data instead of simulated inputs | LiteSense · THDR3K | delta1 / AbsRel: delta1 0.959 -> 0.914; AbsRel 0.054 -> 0.091 | Hurt |
Closest papers by meaning
- MetricAnything: Scaling Metric Depth Pretraining with Noisy Heterogeneous Sources (ECCV 2026 · similarity 0.86)
- PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation (CVPR 2024 · similarity 0.86)
- RDFC-GAN: RGB-Depth Fusion CycleGAN for Indoor Depth Completion (TPAMI 2024 · similarity 0.86)
- Estimating Depth from RGB and Sparse Sensing (ECCV 2018 · similarity 0.86)
- SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation (CVPR 2025 · similarity 0.84)
- UniDepth: Universal 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.
