Research map / Label-efficient and robust learning / Test-time adaptation
ETA: Energy-based Test-time Adaptation for Depth Completion
Younjoon Chung, Hyoungseob Park, Patrick Rim, Xiaoran Zhang, Jihe He, Ziyao Zeng, Safa Cicek, Byung-Woo Hong, James S. Duncan, Alex Wong
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 region-based energy instead of a single image-level energy (ablation) | ETA (MSG-CHN) · KITTI to VKITTI-FOG | MAE: 1.406 -> 0.703 | Improved |
| add region-based energy instead of a single image-level energy (ablation) | ETA (MSG-CHN) · KITTI to VKITTI-FOG | RMSE: 4.226 -> 2.996 | Improved |
| adapt with ETA vs pre-trained model | MSG-CHN · KITTI to VKITTI-FOG | MAE: 2.842 -> 0.703 | Improved |
| use ETA instead of ProxyTTA | CostDCNet · KITTI to Waymo-Hard | MAE: 0.595 -> 0.582 | Improved |
| use ETA instead of ProxyTTA | CostDCNet · KITTI to Waymo-Hard | RMSE: 1.728 -> 1.699 | Improved |
| use ETA instead of ProxyTTA | BP-Net · KITTI to NYUv2 (outdoor-to-indoor) | MAE: 1.380 -> 1.322 | Improved |
| use ETA instead of ProxyTTA | BP-Net · KITTI to SceneNet (outdoor-to-indoor) | MAE: 0.401 -> 0.340 | Improved |
| use ETA instead of ProxyTTA | BP-Net · KITTI to ScanNet (outdoor-to-indoor) | MAE: 0.311 -> 0.272 | Improved |
| use ETA instead of ProxyTTA | MSG-CHN · outdoor targets | MAE: 5.36% | Improved |
| use ETA instead of ProxyTTA | MSG-CHN · outdoor targets | RMSE: 1.97% | Improved |
| use ETA instead of TEA | depth completion models · outdoor targets | MAE: 24.9% | Improved |
| use ETA instead of TEA | depth completion models · outdoor targets | RMSE: 19.0% | Improved |
Closest papers by meaning
- Test-Time Adaptation for Depth Completion (CVPR 2024 · similarity 0.91)
- Zero-shot Depth Completion via Test-time Alignment with Affine-invariant Depth Prior (AAAI 2025 · similarity 0.84)
- PointMAC: Meta-Learned Adaptation for Robust Test-Time Point Cloud Completion (NeurIPS 2025 · similarity 0.84)
- Global and Hierarchical Geometry Consistency Priors for Few-shot NeRFs in Indoor Scenes (CVPR 2024 · similarity 0.84)
- ProtoDepth: Unsupervised Continual Depth Completion with Prototypes (CVPR 2025 · similarity 0.84)
- Depth Anything V2 (NeurIPS 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.
