Research map / 3D vision / Depth and stereo
Structure-Preserving Stereoscopic View Synthesis With Multi-Scale Adversarial Correlation Matching
Yu Zhang, Dongqing Zou, Jimmy S. Ren, Zhe Jiang, Xiaohao Chen
Check 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 MS-ACM to Deep3D | Deep3D · KITTI-Raw | RMSE: 19.466 -> 18.062 | Improved |
| Add MS-ACM to SepConv | SepConv · KITTI-Raw | RMSE: 19.556 -> 18.370 | Improved |
| Add MS-ACM to Deep3D | Deep3D · KITTI-Eigen | RMSE: 22.694 -> 22.159 | Improved |
| Add MS-ACM to SepConv | SepConv · KITTI-Eigen | RMSE: 23.796 -> 23.519 | Improved |
| Add MS-ACM to Deep3D | Deep3D · KITTI-Eigen | PSNR: 21.400 -> 21.624 | Improved |
| Add MS-ACM to SepConv | SepConv · KITTI-Eigen | PSNR: 21.010 -> 21.120 | Improved |
| Use multi-scale rather than single-scale correlation matching (ablation) | MS-ACM · KITTI-Raw | RMSE: 20.870 -> 18.370 | Improved |
| Increase multi-scale correlation window from 3 to 7 (ablation) | MS-ACM · KITTI-Raw | RMSE: 18.370 -> 18.500 | Hurt |
| Increase multi-scale correlation window from 3 to 11 (ablation) | MS-ACM · KITTI-Raw | RMSE: 18.370 -> 18.848 | Hurt |
| Add noise resistance training (ablation) | MS-ACM · KITTI 2015 split | RMSE: 44.662 -> 19.558 | Improved |
| Add feature regularization (ablation) | MS-ACM · KITTI 2015 split | RMSE: 19.558 -> 19.280 | Improved |
| Add self-reconstruction (ablation) | MS-ACM · KITTI 2015 split | RMSE: 19.280 -> 18.370 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Deep3D — Deep3D: Fully Automatic 2D-to-3D Video Conversion with Deep Convolutional Neural Networks (ECCV 2016) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Multi-View Consistent Generative Adversarial Networks for 3D-Aware Image Synthesis (CVPR 2022 · similarity 0.89)
- Single-View View Synthesis with Self-rectified Pseudo-Stereo (IJCV 2023 · similarity 0.88)
- Learning To Render Novel Views From Wide-Baseline Stereo Pairs (CVPR 2023 · similarity 0.88)
- "Smoothness, Synthesis, and Sampling: Re-thinking Unsupervised Multi-View Stereo with DIV Loss" (ECCV 2024 · similarity 0.87)
- Generalizable Novel-View Synthesis using a Stereo Camera (CVPR 2024 · similarity 0.87)
- Geometry-guided Online 3D Video Synthesis with Multi-View Temporal Consistency (CVPR 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.
