Research map / Architectures and efficiency / RNNs and normalization
Holistically-Attracted Wireframe Parsing
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 HAWP instead of re-trained L-CNN | Wireframe parsing · Wireframe | msAP: 62.9 -> 65.7; 2.8% absolute improvement | Helped |
| Use HAWP instead of re-trained L-CNN | Wireframe parsing · YorkUrban | msAP: 26.8 -> 28.1; 1.3% absolute improvement | Helped |
| Use HAWP instead of re-trained L-CNN | Wireframe parsing · Wireframe | FPS: 15.6 -> 29.5; 89% relative improvement | Helped |
| Add the full set of D*, D-, and S- negative samplers alongside natural negatives (ablation) | H-AFM verification training · Wireframe | sAP^10: 66.0 -> 66.5 | Helped |
| Use natural negatives instead of D*, D-, and S- samplers (ablation) | H-AFM verification training · Wireframe | sAP^10: 66.1 -> 66.5 | Helped |
| Add natural negatives to the D- and S- samplers (ablation) | H-AFM verification training · Wireframe | sAP^10: 65.8 -> 66.5 | Helped |
| Add distance residual correction at testing (ablation) | H-AFM line proposal generation · Wireframe | sAP^10: 63.0 -> 66.5 | Helped |
| Add distance residual training and testing (ablation) | H-AFM line proposal generation · Wireframe | sAP^10: 62.6 -> 66.5 | Helped |
| Replace vanilla AFM with H-AFM (ablation) | HAWP line proposal generation · Wireframe | sAP^10: 33.7 -> 66.5 | Helped |
Closest papers by meaning
- Learning to Parse Wireframes in Images of Man-Made Environments (CVPR 2018 · similarity 0.84)
- NEAT: Distilling 3D Wireframes from Neural Attraction Fields (CVPR 2024 · similarity 0.83)
- Hierarchy-Agnostic Unsupervised Segmentation: Parsing Semantic Image Structure (NeurIPS 2024 · similarity 0.81)
- Spot the Error: Non-autoregressive Graphic Layout Generation with Wireframe Locator (AAAI 2024 · similarity 0.80)
- Data-independent Module-aware Pruning for Hierarchical Vision Transformers (ICLR 2024 · similarity 0.79)
- Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields (CVPR 2024 · similarity 0.79)
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.