Research map / Probabilistic and scientific modeling / Variational and generative sampling
Hierarchical Discrete Distribution Decomposition for Match Density 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 HD³F match probabilities instead of PWCNet vectorized flow for propagation | semantic region propagation · KITTI multiview extension | mean IoU: 38.3% -> 39.7% | Helped |
| use HD³F match probabilities instead of PWCNet vectorized flow for propagation | semantic region propagation · KITTI multiview extension | mean Acc: 60.5% -> 63.6% | Helped |
| classify predictions using the HD³F uncertainty map instead of forward-backward consistency | HD³F · KITTI 2015 training, Noc, Mean | IoU: 50.9 -> 66.9 | Helped |
| classify predictions using the HD³F uncertainty map instead of forward-backward consistency | HD³F · KITTI 2015 training, Noc, Mean | Acc: 75.4 -> 77.8 | Helped |
| classify predictions using the HD³F uncertainty map instead of forward-backward consistency | HD³F · KITTI 2015 training, All, Mean | IoU: 49.5 -> 68.0 | Helped |
| classify predictions using the HD³F uncertainty map instead of forward-backward consistency | HD³F · KITTI 2015 training, All, Mean | Acc: 79.4 -> 85.1 | Helped |
| replace DLA with the deeper DLA-Up feature encoder (ablation) | HD³S · KITTI 2012 stereo test, Out-Noc | error rate: 1.83% -> 1.68% | Helped |
| replace DLA with the deeper DLA-Up feature encoder (ablation) | HD³S · KITTI 2012 stereo test, Out-All | error rate: 2.43% -> 2.25% | Helped |
| replace DLA with the deeper DLA-Up feature encoder (ablation) | HD³S · KITTI 2015 stereo test, D1-bg | error rate: 2.12% -> 2.06% | Helped |
| replace DLA with the deeper DLA-Up feature encoder (ablation) | HD³S · KITTI 2015 stereo test, D1-fg | error rate: 5.40% -> 5.47% | Hurt |
| replace DLA with the deeper DLA-Up feature encoder (ablation) | HD³S · KITTI 2015 stereo test, D1-all | error rate: 2.67% -> 2.63% | Helped |
Closest papers by meaning
- UFM: A Simple Path towards Unified Dense Correspondence with Flow (NeurIPS 2025 · similarity 0.86)
- Diffusion Model for Dense Matching (ICLR 2024 · similarity 0.85)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.85)
- Adapting Dense Matching for Homography Estimation with Grid-based Acceleration (CVPR 2025 · similarity 0.84)
- Hierarchical Object-Aware Dual-Level Contrastive Learning for Domain Generalized Stereo Matching (NeurIPS 2024 · similarity 0.84)
- Discrete Flow Matching (NeurIPS 2024 · similarity 0.84)
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.