Research map / Architectures and efficiency / RNNs and normalization
AANet: Adaptive Aggregation Network for Efficient Stereo Matching
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 ISA and CSA modules (ablation) | AANet cost aggregation · KITTI 2015 validation | EPE: 0.75 -> 0.68 | Helped |
| Add ISA and CSA modules (ablation) | AANet cost aggregation · KITTI 2015 validation | D1-all: 2.63 -> 2.29 | Helped |
| Add ISA (ablation) | AANet cost aggregation without ISA · KITTI 2015 validation | EPE: 0.70 -> 0.68 | Helped |
| Add ISA (ablation) | AANet cost aggregation without ISA · KITTI 2015 validation | D1-all: 2.22 -> 2.29 | Mixed |
| Add CSA (ablation) | AANet cost aggregation without CSA · KITTI 2015 validation | EPE: 0.69 -> 0.68 | Helped |
| Add CSA (ablation) | AANet cost aggregation without CSA · KITTI 2015 validation | D1-all: 2.31 -> 2.29 | Hurt |
| Use GA-Net predictions as pseudo-labels for pixels without ground-truth disparities | AANet · KITTI 2015 validation | D1-all: 2.29 to 2.15 | Helped |
| Use GA-Net predictions as pseudo-labels for pixels without ground-truth disparities | AANet · KITTI 2015 validation | EPE: 0.68 to 0.69 | Hurt |
| Replace cost-aggregation 3D convolutions with AANet adaptive aggregation | GC-Net · Scene Flow test | EPE: 2.51 -> 0.98 | Helped |
| Replace cost-aggregation 3D convolutions with AANet adaptive aggregation | PSMNet · Scene Flow test | EPE: 1.09 -> 0.97 | Helped |
| Replace cost-aggregation 3D convolutions with AANet adaptive aggregation | GA-Net · Scene Flow test | EPE: 0.84 -> 0.87 | Hurt |
| Replace cost-aggregation 3D convolutions with AANet adaptive aggregation | StereoNet · Scene Flow test | EPE: 1.10 -> 1.08 | Helped |
Closest papers by meaning
- GA-Net: Guided Aggregation Net for End-To-End Stereo Matching (CVPR 2019 · similarity 0.93)
- Deep Stereo Matching With Explicit Cost Aggregation Sub-Architecture (AAAI 2018 · similarity 0.93)
- CFNet: Cascade and Fused Cost Volume for Robust Stereo Matching (CVPR 2021 · similarity 0.91)
- Improved Stereo Matching With Constant Highway Networks and Reflective Confidence Learning (CVPR 2017 · similarity 0.90)
- Any-Stereo: Arbitrary Scale Disparity Estimation for Iterative Stereo Matching (AAAI 2024 · similarity 0.90)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.89)
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.