Research map / 3D vision / Depth and stereo
RayMVSNet: Learning Ray-Based 1D Implicit Fields for Accurate Multi-View Stereo
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 |
|---|---|---|---|
| add the epipolar transformer (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.343 -> 0.330 | Helped |
| add 2D image features (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.348 -> 0.330 | Helped |
| add 3D volume features (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.378 -> 0.330 | Helped |
| use epipolar transformer instead of visibility-aware max-pooling aggregation (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.338 -> 0.330 | Helped |
| use ray-based local fields instead of one global implicit field (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.608 -> 0.330 | Helped |
| use LSTM instead of Transformer for sequential ray modeling (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.341 -> 0.330 | Helped |
| use LSTM instead of average pooling for ray aggregation (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.381 -> 0.330 | Helped |
| use LSTM instead of max pooling for ray aggregation (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.424 -> 0.330 | Helped |
| add auxiliary SDF prediction (ablation) | RayMVSNet · DTU test | Overall distance (mm): 0.342 -> 0.330 | Helped |
| add moderate Gaussian perturbation to coarse depth during training and testing | RayMVSNet · DTU test | depth prediction quality | Helped |
Closest papers by meaning
- Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement (CVPR 2020 · similarity 0.90)
- Multi-View Stereo Representation Revist: Region-Aware MVSNet (CVPR 2023 · similarity 0.90)
- MVSFormer++: Revealing the Devil in Transformer's Details for Multi-View Stereo (ICLR 2024 · similarity 0.89)
- A Global Depth-Range-Free Multi-View Stereo Transformer Network with Pose Embedding (NeurIPS 2024 · similarity 0.88)
- Zero-Shot Novel View and Depth Synthesis with Multi-View Geometric Diffusion (CVPR 2025 · similarity 0.88)
- RRT-MVS: Recurrent Regularization Transformer for Multi-View Stereo (AAAI 2025 · similarity 0.87)
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.