From Elevation Maps To Contour Lines: SVM and Decision Trees to Detect Violin Width Reduction
arXiv cs.AI / 4/6/2026
💬 OpinionIdeas & Deep AnalysisModels & Research
Key Points
- The study investigates automatic detection of violin width reduction using 3D photogrammetric meshes as the underlying data source.
- It compares two machine-learning approaches—SVM and Decision Trees—fed by either a raw geometry representation derived from elevation maps or a feature-engineered representation based on fitted parametric contour lines.
- While elevation-map-based inputs sometimes produce strong performance, they generally fail to outperform the contour-based, targeted features.
- The results suggest that contour-line fitting provides a more effective geometric signal for this specific shape-change detection task than more generic elevation-map representations.
