Research map / Detection and segmentation / CNN features
CVNet: Contour Vibration Network for Building Extraction
Ziqiang Xu, Chunyan Xu, Zhen Cui, Xiangwei Zheng, Jian Yang
AI-extracted. The measured effects below were read by AI from the paper's full text (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 CVNet contour vibration evolution instead of TDAC | Building extraction · Vaihingen | mIoU: 89.16 -> 90.43 | Improved |
| Use CVNet contour vibration evolution instead of TDAC | Building extraction · Vaihingen | BoundF: 78.12 -> 81.76 | Improved |
| Use CVNet contour vibration evolution instead of TDAC | Building extraction · Bing Huts | mIoU: 80.39 -> 80.42 | No clear effect |
| Use CVNet contour vibration evolution instead of TDAC | Building extraction · Bing Huts | BoundF: 53.50 -> 46.37 | Hurt |
| Use CVNet contour vibration evolution instead of DARNet | Building extraction · Inria-building | mIoU: 65.83 -> 77.61 | Improved |
| Use CVNet contour vibration evolution instead of DARNet | Building extraction · Inria-building | BoundF: 33.04 -> 42.20 | Improved |
| Increase the number of contour points up to 60 (ablation) | CVNet · Vaihingen and Bing Huts | mIoU | Improved |
| Increase contour points beyond 60 (ablation) | CVNet · Vaihingen and Bing Huts | mIoU | Hurt |
| Increase contour evolution steps up to 70 (ablation) | CVNet · Vaihingen and Bing Huts | mIoU | Improved |
| Increase contour evolution steps beyond 70 (ablation) | CVNet · Vaihingen and Bing Huts | mIoU | Hurt |
Closest papers by meaning
- A Generalized Contour Vibration Model for Building Extraction (IJCV 2025 · similarity 0.85)
- TopoRefine: Plug-and-Play Topology-Aware Contour Refinement for Building Segmentation (NeurIPS 2026 · similarity 0.84)
- Conv-MPN: Convolutional Message Passing Neural Network for Structured Outdoor Architecture Reconstruction (CVPR 2020 · similarity 0.84)
- DARNet: Deep Active Ray Network for Building Segmentation (CVPR 2019 · similarity 0.84)
- Learning Deep Structured Active Contours End-to-End (CVPR 2018 · similarity 0.84)
- Learning to Predict Structural Vibrations (NeurIPS 2024 · similarity 0.81)
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.
