Research map / Detection and segmentation / CNN features
Learning Deep Structured Active Contours End-to-End
Read the paperCheck 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 |
|---|---|---|---|
| Replace the CNN segmentation baseline with DSAC active-contour refinement | Building footprint segmentation with manual initialization · Vaihingen test | Average IoU: 0.78 -> 0.84 | Helped |
| Replace the CNN segmentation baseline with DSAC active-contour refinement | Building footprint segmentation with manual initialization · Bing huts test | Average IoU: 0.56 -> 0.65 | Helped |
| Replace the CNN segmentation baseline with DSAC active-contour refinement | Building footprint segmentation with manual initialization · Bing huts test | RMSE (m^2): 23.9 -> 13.4 | Helped |
| Apply DSAC initialized with raw DWT polygons | DWT raw · TorontoCity validation | WeighCov: 0.42 -> 0.55 | Helped |
| Apply DSAC initialized with post-processed DWT polygons | DWT postproc. · TorontoCity validation | WeighCov: 0.52 -> 0.57 | Helped |
| Train with raw DWT polygons and test with post-processed DWT polygons (ablation) | DSAC · TorontoCity validation | WeighCov: 0.57 -> 0.58 | Helped |
| Train with raw DWT polygons and test with post-processed DWT polygons (ablation) | DSAC · TorontoCity validation | PolySim: 0.26 -> 0.27 | Helped |
| Use local rather than scalar kappa and beta (ablation) | DSAC · Vaihingen test | Average IoU: 0.64 -> 0.84 | Helped |
| Use local rather than scalar kappa and beta (ablation) | DSAC · Bing huts test | Average IoU: 0.60 -> 0.65 | Helped |
| Add the balloon term kappa (ablation) | DSAC · Vaihingen test | Average IoU: 0.63 -> 0.84 | Helped |
| Add the balloon term kappa (ablation) | DSAC · Bing huts test | Average IoU: 0.42 -> 0.65 | Helped |
| Predict local rather than scalar alpha (ablation) | DSAC · Vaihingen test | Average IoU: 0.84 -> 0.83 | No clear effect |
Closest papers by meaning
- DARNet: Deep Active Ray Network for Building Segmentation (CVPR 2019 · similarity 0.88)
- Building Instance Segmentation for Dense Urban Settlements (AAAI 2026 · similarity 0.86)
- Joint Semantic-geometric Learning for Polygonal Building Segmentation (AAAI 2021 · similarity 0.85)
- TopoRefine: Plug-and-Play Topology-Aware Contour Refinement for Building Segmentation (NeurIPS 2026 · similarity 0.85)
- Boundary-Aware 3D Building Reconstruction From a Single Overhead Image (CVPR 2020 · similarity 0.85)
- Bundle Pooling for Polygonal Architecture Segmentation Problem (CVPR 2020 · 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.