Research map / 3D vision / Depth and stereo
Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace a single uniform plane-sweep volume with cascaded uncertainty-guided adaptive thin volumes | UCS-Net · DTU testing set | Overall: 0.376 -> 0.344 | Helped |
| Add stage 2 refinement after stage 1 (ablation) | UCS-Net · DTU testing set | Overall: 0.539 -> 0.399 | Helped |
| Add stage 3 refinement after stages 1–2 (ablation) | UCS-Net · DTU testing set | Overall: 0.399 -> 0.344 | Helped |
| Bilinearly upsample stage 1 output by x2 instead of using learned ATV refinement (ablation) | UCS-Net · DTU testing set | Overall: 0.344 -> 0.473 | Hurt |
| Bilinearly upsample stage 2 output by x2 instead of using learned ATV refinement (ablation) | UCS-Net · DTU testing set | Overall: 0.344 -> 0.364 | Hurt |
| Use a learned multi-stage method rather than Dense-R-MVSNet | UCS-Net · Tanks and Temples intermediate | Mean F-score: 50.55 -> 54.83 | Helped |
| Use the full three-stage reconstruction rather than MVSNet | UCS-Net · DTU validate set | Running time: 1.049 -> 0.257 | Helped |
| Use the full three-stage reconstruction rather than MVSNet | UCS-Net · DTU validate set | Memory: 4511 -> 1647 | Helped |
Closest papers by meaning
- Cost Volume Pyramid Based Depth Inference for Multi-View Stereo (CVPR 2020 · similarity 0.90)
- Uncertainty-Aware Deep Multi-View Photometric Stereo (CVPR 2022 · similarity 0.90)
- Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation (AAAI 2023 · similarity 0.90)
- MVS2D: Efficient Multi-View Stereo via Attention-Driven 2D Convolutions (CVPR 2022 · similarity 0.90)
- GeoMVSNet: Learning Multi-View Stereo With Geometry Perception (CVPR 2023 · similarity 0.89)
- A Global Depth-Range-Free Multi-View Stereo Transformer Network with Pose Embedding (NeurIPS 2024 · similarity 0.88)
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.