Research map / Video / Video segmentation and tracking
SpatioTemporal Difference Network for Video Depth Super-Resolution
Zhengxue Wang, Yuan Wu, Xiang Li, Zhiqiang Yan, Jian Yang
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 spatial difference (SD) instead of concatenation (ablation) | STDNet · DyDToF (x4) | RMSE: reduce RMSE by 3.56cm | Improved |
| add temporal difference (TD) instead of concatenation (ablation) | STDNet · DyDToF (x4) | RMSE: reduce RMSE by 14.02cm | Improved |
| combine SD and TD (ablation) | STDNet · DyDToF (x4) | RMSE: surpassing the baseline by 17.94cm | Improved |
| add spatial and temporal difference losses to reconstruction loss (ablation) | STDNet · TarTanAir (x16) | RMSE: surpasses the baseline by 7.08cm in RMSE | Improved |
| use more than 2 neighboring frames in TD (ablation) | STDNet · TarTanAir (x16) | RMSE | No clear effect |
| use multiple neighboring frames instead of one RGB-D frame (ablation) | STDNet · TarTanAir (x16) | RMSE | Improved |
| replace single-frame methods with STDNet (complexity) | STDNet · TarTanAir (x8) | RMSE: average reduction of 9.23M parameters and 35.82cm RMSE | Improved |
| STDNet vs multi-frame DVSR | STDNet · TarTanAir (x8) | inference time: 47.35ms improvement in inference speed and a 4.93cm gain | Improved |
Closest papers by meaning
- Real-Time Video Super-Resolution With Spatio-Temporal Networks and Motion Compensation (CVPR 2017 · similarity 0.89)
- Look Back and Forth: Video Super-Resolution With Explicit Temporal Difference Modeling (CVPR 2022 · similarity 0.89)
- Fast Spatio-Temporal Residual Network for Video Super-Resolution (CVPR 2019 · similarity 0.88)
- SGNet: Structure Guided Network via Gradient-Frequency Awareness for Depth Map Super-resolution (AAAI 2024 · similarity 0.88)
- Video Depth Anything: Consistent Depth Estimation for Super-Long Videos (CVPR 2025 · similarity 0.86)
- Sequence Matters: Harnessing Video Models in 3D Super-Resolution (AAAI 2025 · 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.
