Research map / 3D vision / Face and shape
A Condition Number for Joint Optimization of Cycle-Consistent Networks
Leonidas Guibas, Qixing Huang, Zhenxiao Liang
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| add SDP-optimized cycle weights instead of uniform weights (ablation) | joint functional map optimization · ShapeCoSeg | CD % of correspondences: -5.2% CD at egeo=0.15 with uniform weights | Improved |
| run 10x more iterations with uniform cycle weights (ablation) | joint functional map optimization with uniform weights · ShapeCoSeg | CD % of correspondences: gap 5.2% -> 2.4% | No clear effect |
| add SDP-optimized cycle weights instead of uniform weights (ablation) | cycle-consistent flow networks · PASCAL3D | CD % of correspondences: -4.3% CD with uniform weights | Improved |
| allow per-edge network parameters instead of identical parameters (ablation) | cycle-consistent flow networks · PASCAL3D | CD % of correspondences: -3.2% CD with identical parameters | Improved |
| use exact cycle-consistency with weighted cycles instead of directed path-invariance (Zhang19) | joint shape matching · ShapeCoSeg | CD % at egeo=0.1: +4.9% | Improved |
| add cycle-consistency on real images to synthetic-only training | dense image flow network · PASCAL3D | CD % at eeuc=0.1: +8.1% | Improved |
Closest papers by meaning
- Multi-Shape Matching with Cycle Consistency Basis via Functional Maps (AAAI 2025 · similarity 0.87)
- $C^2M^3$: Cycle-Consistent Multi-Model Merging (NeurIPS 2024 · similarity 0.87)
- Higher-Order Ratio Cycles for Fast and Globally Optimal Shape Matching (CVPR 2025 · similarity 0.86)
- Incremental Topological Ordering and Cycle Detection with Predictions (ICML 2024 · similarity 0.85)
- Are Graph Neural Networks Optimal Approximation Algorithms? (NeurIPS 2024 · similarity 0.84)
- Discrete Cycle-Consistency Based Unsupervised Deep Graph Matching (AAAI 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.
