Research map / Architectures and efficiency / CNN design
LAF-Net: Locally Adaptive Fusion Networks for Stereo Confidence Estimation
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 tri-modal input: matching cost, disparity, and color image (ablation) | LAF-Net · MID 2006, MID 2014, KITTI 2015 | AUC: 0.0364; 0.0683; 0.0225 | Helped |
| Use locally varying attention to fuse modality features (ablation) | LAF-Net · MID 2006, MID 2014, KITTI 2015 | AUC: 0.0386 -> 0.0364; 0.0689 -> 0.0683; 0.0238 -> 0.0225 | Helped |
| Use locally varying scale inference (ablation) | LAF-Net · MID 2006, MID 2014, KITTI 2015 | AUC: 0.0371 -> 0.0364; 0.0685 -> 0.0683; 0.0229 -> 0.0225 | Helped |
| Use recursive confidence refinement (ablation) | LAF-Net · MID 2006, MID 2014, KITTI 2015 | AUC: 0.0372 -> 0.0364; 0.0685 -> 0.0683; 0.0231 -> 0.0225 | Helped |
| Use LAF-Net rather than LGC-Net | Stereo confidence estimation · MID 2006, MID 2014, KITTI 2015 | AUC: 0.0413 -> 0.0405; 0.0389 -> 0.0364; 0.0735 -> 0.0718; 0.0685 -> 0.0683; 0.0392 -> 0.0385; 0.0236 -> 0.0225 | Helped |
Closest papers by meaning
- Learning to Predict Stereo Reliability Enforcing Local Consistency of Confidence Maps (CVPR 2017 · similarity 0.90)
- Modeling Stereo-Confidence out of the End-to-End Stereo-Matching Network via Disparity Plane Sweep (AAAI 2024 · similarity 0.89)
- Real-Time Self-Adaptive Deep Stereo (CVPR 2019 · similarity 0.87)
- Improved Stereo Matching With Constant Highway Networks and Reflective Confidence Learning (CVPR 2017 · similarity 0.86)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.85)
- IINet: Implicit Intra-inter Information Fusion for Real-Time Stereo Matching (AAAI 2024 · similarity 0.85)
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.