Research map / Label-efficient and robust learning / Domain adaptation and generalization
S2R-DepthNet: Learning a Generalizable Depth-Specific Structural Representation
Xiaotian Chen, Yuwang Wang, Xuejin Chen, Wenjun Zeng
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 STE to the raw-image depth baseline (ablation) | S2R-DepthNet · vKITTI→KITTI, 0~80m | Abs Rel: 0.236 -> 0.201 | Improved |
| add DSA to the STE-based model (ablation) | S2R-DepthNet · vKITTI→KITTI, 0~80m | Abs Rel: 0.201 -> 0.165 | Improved |
| combine the attention map with structure by element-wise multiplication instead of addition (ablation) | STE + DSA depth model · vKITTI→KITTI, 0~80m | Abs Rel: 0.185 -> 0.165 | Improved |
| combine the attention map with structure by element-wise multiplication instead of concatenation (ablation) | STE + DSA depth model · vKITTI→KITTI, 0~80m | Abs Rel: 0.179 -> 0.165 | Improved |
| use Sobel edge maps instead of the learned depth-specific representation (ablation) | depth estimation baseline · vKITTI→KITTI, 0~80m | Abs Rel: 0.165 -> 0.217 | Hurt |
| use semantic segmentation maps instead of the learned depth-specific representation (ablation) | depth estimation baseline · vKITTI→KITTI, 0~80m | Abs Rel: 0.165 -> 0.204 | Hurt |
| add STE to the raw-image baseline (ablation) | S2R-DepthNet · SUNCG→NYU Depth v2 | Abs Rel: 0.278 -> 0.225 | Improved |
| add DSA to the STE-based model (ablation) | S2R-DepthNet · SUNCG→NYU Depth v2 | Abs Rel: 0.225 -> 0.196 | Improved |
Closest papers by meaning
- S³Net: Semantic-Aware Self-supervised Depth Estimation with Monocular Videos and Synthetic Data (ECCV 2020 · similarity 0.87)
- Geometry-Aware Symmetric Domain Adaptation for Monocular Depth Estimation (CVPR 2019 · similarity 0.87)
- Learning Scene Structure Guidance via Cross-Task Knowledge Transfer for Single Depth Super-Resolution (CVPR 2021 · similarity 0.87)
- Learning to Predict 3D Surfaces of Sculptures from Single and Multiple Views (IJCV 2018 · similarity 0.86)
- Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation (CVPR 2020 · similarity 0.86)
- Large Spatial Model: End-to-end Unposed Images to Semantic 3D (NeurIPS 2024 · similarity 0.86)
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.
