Research map / Image generation / Super-resolution
Learning Parallax Attention for Stereo Image Super-Resolution
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 |
|---|---|---|---|
| Use stereo input with parallax attention instead of a single left image (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.27 -> 25.43 | Helped |
| Use stereo input rather than replicated left-image inputs (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.29 -> 25.43 | Helped |
| Add residual connections to the ASPP module (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.40 -> 25.43 | Helped |
| Use atrous rather than ordinary convolutions in the ASPP module (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.38 -> 25.43 | Helped |
| Add parallax-attention correspondence aggregation (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.28 -> 25.43 | Helped |
| Add a transition residual block to PAM (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.36 -> 25.43 | Helped |
| Use PAM instead of a 4D cost volume with two 3D convolutional layers (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.23 -> 25.43 | Helped |
| Use PAM instead of a 4D cost volume with two 3D convolutional layers (ablation) | PASSRnet · KITTI 2015 ×4 | running time: 1.5 times | Helped |
| Add photometric loss to SR loss (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.35 -> 25.38 | Helped |
| Add smoothness loss to SR and photometric losses (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.38 -> 25.40 | Helped |
| Add cycle loss to SR, photometric, and smoothness losses (ablation) | PASSRnet · KITTI 2015 ×4 | PSNR: 25.40 -> 25.43 | Helped |
| Use PASSRnet rather than the strongest listed single-image SR baseline, DRRN | Stereo SR · Middlebury ×2 | PSNR: +1.04 dB | Helped |
Closest papers by meaning
- Enhancing the Spatial Resolution of Stereo Images Using a Parallax Prior (CVPR 2018 · similarity 0.90)
- Learned Binocular-Encoding Optics for RGBD Imaging Using Joint Stereo and Focus Cues (CVPR 2025 · similarity 0.86)
- Selective-Stereo: Adaptive Frequency Information Selection for Stereo Matching (CVPR 2024 · similarity 0.85)
- CFAT: Unleashing Triangular Windows for Image Super-resolution (CVPR 2024 · similarity 0.84)
- Any-Stereo: Arbitrary Scale Disparity Estimation for Iterative Stereo Matching (AAAI 2024 · similarity 0.84)
- DORNet: A Degradation Oriented and Regularized Network for Blind Depth Super-Resolution (CVPR 2025 · similarity 0.83)
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.