arxiv: 2604.12631 · v1 · submitted 2026-04-14 · 💻 cs.CG

Recognition: unknown

Topology Understanding of B-Spline Surface/Surface Intersection with Mapper

Chenming Gao, Gengchen Li, Hongwei Lin

Authors on Pith no claims yet

Pith reviewed 2026-05-10 14:00 UTC · model grok-4.3

classification 💻 cs.CG

keywords B-spline surfacessurface intersectiontopologyMappertopological data analysisCADintersection curvesingularities

0 comments

The pith

Mapper recovers the topology of B-spline surface intersections from sampled points.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a Mapper-based procedure that takes point samples along the intersection of two B-spline surfaces and builds a topological summary of the result. Subdivision algorithms already compute the geometric intersection curve but often leave its connectivity, loops, and singularities unclassified. The new method aims to resolve those ambiguities for both ordinary and degenerate cases. Readers would care because accurate topology determines whether subsequent CAD operations such as trimming or boolean unions succeed or produce invalid models.

Core claim

Sampling points on the intersection curve of two B-spline surfaces and processing them with the Mapper algorithm produces a graph or complex whose connected components and cycles correctly reproduce the topology of the intersection, including cases with self-intersections, pinch points, and multiple loops.

What carries the argument

The Mapper algorithm, which builds a topological summary by covering a point cloud with overlapping sets, clustering within each set, and connecting clusters that share points across overlaps.

If this is right

The method classifies both ordinary and complex intersection topologies that subdivision alone cannot resolve.
Experimental tests confirm robustness and topological correctness across the chosen examples.
The approach can be inserted after any subdivision-based intersection routine to supply the missing topology.
It supports cases with degenerate components without requiring manual classification.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Existing CAD intersection pipelines could adopt the Mapper post-processing step to reduce downstream modeling failures.
Adaptive sampling guided by local curvature might be needed to keep the point cloud size practical for high-degree surfaces.
The same Mapper pipeline could be tested on intersections involving other parametric surfaces or higher-dimensional objects.

Load-bearing premise

A finite point sample drawn from the intersection curve is dense enough for Mapper to reconstruct the correct global connectivity even when the curve contains singularities or self-intersections.

What would settle it

Apply the method to a pair of B-spline surfaces known to intersect in a single figure-eight self-crossing curve and check whether the Mapper output graph reports one connected component with a crossing rather than two separate loops.

Figures

Figures reproduced from arXiv: 2604.12631 by Chenming Gao, Gengchen Li, Hongwei Lin.

**Figure 1.** Figure 1: An example of the Mapper algorithm. The data is sampled from a noisy circle, and the filter function is 𝑓(𝑥) = 𝑥1 , where 𝑥 = (𝑥1 , 𝑥2 ) is a point in 𝑋. (a) The point set 𝑋 colored by the value of the filter function 𝑓. (b)The range of 𝑓 is covered with five equal-length intervals, with a 20% overlap of neighboring intervals. (c)The DBSCAN algorithm [Ester et al., 1996] is used to cluster the preimage for… view at source ↗

**Figure 3.** Figure 3: Generation of the initial Mapper graph, 𝜃ov = 0.2. (a)Input point set 𝑋. (b)𝑋 is colored according to the values of the filter function 𝑓. The black dashed line passes through the center of 𝑋, indicating the principal direction of 𝑋. (c)Cover construction. Only a subset of intervals is visualized for clarity. Points corresponding to the same interval are enclosed between dashed lines, and the preimages of … view at source ↗

**Figure 4.** Figure 4: An illustration of additional cross-graph edges introduced by independent subdivision of adjacent Mapper nodes. Top: data elements corresponding to two adjacent nodes, where node-specific elements are shown in red and blue, and their shared elements are shown in green. Middle: the corresponding two adjacent Mapper nodes and their connecting edge. Bottom: Mapper subgraphs generated independently from the … view at source ↗

**Figure 5.** Figure 5: An example of intersecting set partition based on the Mapper graph. (a) The intersection point set 𝑃1,𝑢,𝑣 (b) 𝜃ov = 0.1 (c) 𝜃ov = 0.2 (d) 𝜃ov = 0.3 (e) 𝜃ov = 0.4 [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗

**Figure 6.** Figure 6: An example of the impact of changing the overlap ratio parameter 𝜃ov. and Hess, 2021] to compute the Mapper graph. In our experiments, we will show that the algorithm can successfully understand the topology of intersection. 5.1. Parameter selection for the Two-step Mapper algorithm The overlap ratio 𝜃ov is a key parameter in the Mapper algorithm. This section analyzes its influence on the resulting grap… view at source ↗

read the original abstract

In the realm of computer-aided design (CAD) software, the intersection of B-spline surfaces stands as a fundamental operation. Despite the extensive history of surface intersection algorithms, the challenge of handling complex intersection topologies persists. While subdivision algorithms have demonstrated strong robustness in computing surface/surface intersection and are capable of addressing singular cases, determining the topology of the intersection obtained through these methods is a key factor for calculating correct intersection, and remains a difficult issue. To address this challenge, we propose a Mapper-based method for determining the topology of the intersection between two B-spline surfaces. Our algorithm is designed to efficiently handle various common and complex intersection topologies. Experimental results verify the robustness and topological correctness of this method.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Mapper on B-spline intersection point clouds is a new angle for topology recovery, but the abstract gives no evidence it handles singularities reliably.

read the letter

The paper's main contribution is applying the Mapper algorithm to a point cloud obtained from B-spline surface intersections in order to recover the topology of the resulting curve. This combination has not been tried before in the CAD literature, so the framing as a response to the topology bottleneck after subdivision is genuinely new. The authors correctly note that subdivision methods are good at producing points even in singular cases, but turning those points into a correct topological description remains difficult. That observation is fair and worth pursuing. What the work does well is keep the focus on a practical engineering problem rather than abstract topology for its own sake. The algorithm is presented as efficient for both common and complex topologies, which aligns with daily needs in surface modeling. The soft spot is that the central claim rests on an unexamined transfer from discrete samples to exact topology. Mapper's output depends on the filter function, the cover, and the overlap parameter; nothing in the description shows how these are chosen to avoid collapsing distinct branches or misclassifying crossings at cusps and self-intersections. The abstract states that experiments confirm robustness and correctness, yet supplies no test cases, sampling densities, singular configurations, or baseline comparisons. Without those, it is impossible to tell whether the method actually improves on existing topology classification techniques or simply reproduces their failures on the hard cases. This paper is for people working on surface intersection algorithms in computational geometry and CAD. A reader already familiar with subdivision and TDA might see a useful direction to explore. It deserves peer review so that referees can examine the implementation details, the choice of Mapper parameters, and the actual experimental evidence. The idea is narrow but concrete enough to justify that step.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes a Mapper-based algorithm to recover the topology of the intersection curve between two B-spline surfaces. The approach first obtains an intersection point cloud (via subdivision) and then applies the Mapper construction from topological data analysis to extract connectivity and higher invariants, with the claim that the method efficiently handles both standard and complex topologies including singularities.

Significance. If the transfer from discrete samples to exact algebraic topology is reliable, the work would address a long-standing gap in CAD: subdivision methods produce robust point sets but leave topology classification (number of components, crossings, cusps) unresolved. Combining geometric modeling with TDA offers a fresh direction, yet the absence of any described validation leaves the practical impact speculative.

major comments (2)

Abstract: the assertion that 'Experimental results verify the robustness and topological correctness of this method' is unsupported by any mention of test cases, sampling density, filter functions, cover parameters, comparison baselines, or singular configurations examined. Without these, the central claim cannot be assessed.
Method description (inferred from abstract): Mapper applied to an intersection point cloud recovers topology only conditionally on filter choice, cover resolution, and overlap; the manuscript provides no argument or guarantee that the output simplicial complex matches the topology of the real algebraic curve when the intersection contains self-intersections, cusps, or degenerate components, where discrete samples do not encode crossing information.

minor comments (1)

The abstract should briefly indicate how the initial intersection point cloud is generated (e.g., specific subdivision tolerances or marching methods) to clarify the pipeline.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback, which has helped us improve the clarity and completeness of our presentation. We address each major comment below and have made corresponding revisions to the manuscript.

read point-by-point responses

Referee: Abstract: the assertion that 'Experimental results verify the robustness and topological correctness of this method' is unsupported by any mention of test cases, sampling density, filter functions, cover parameters, comparison baselines, or singular configurations examined. Without these, the central claim cannot be assessed.

Authors: We agree that the abstract claim would be stronger with supporting details. In the revised manuscript we have expanded the abstract to reference the specific test cases (including standard loops, self-intersections, cusps, and degenerate components), the subdivision sampling densities employed, the filter functions and cover parameters used in the Mapper construction, and the baseline topology-recovery methods against which we compared. A new experimental section now supplies the full parameter tables, quantitative correctness metrics, and singular-configuration results that were previously only summarized. revision: yes
Referee: Method description (inferred from abstract): Mapper applied to an intersection point cloud recovers topology only conditionally on filter choice, cover resolution, and overlap; the manuscript provides no argument or guarantee that the output simplicial complex matches the topology of the real algebraic curve when the intersection contains self-intersections, cusps, or degenerate components, where discrete samples do not encode crossing information.

Authors: The referee correctly notes the parameter dependence of Mapper. We have revised the method section to specify our filter-function selection (projection onto a coordinate aligned with the dominant curvature of the intersection curve) and the adaptive cover-resolution strategy that ensures sufficient overlap to capture connectivity. We now include a short argument showing that, for point clouds generated by subdivision at a density exceeding the local feature size, the resulting nerve complex recovers the correct number of components and loops; self-intersections are detected as multiple overlapping simplices. For cusps and degeneracies we acknowledge that discrete samples alone cannot encode crossing multiplicity, and we have added an explicit limitations paragraph together with empirical examples illustrating when the method succeeds and when further geometric refinement is required. revision: partial

Circularity Check

0 steps flagged

No circularity: algorithmic Mapper procedure with no derivations or self-referential reductions

full rationale

The paper presents a computational algorithm that samples the intersection curve of two B-spline surfaces and applies the Mapper algorithm to recover its topology. The provided abstract and description contain no equations, no fitted parameters, no predictions derived from inputs by construction, and no load-bearing self-citations. The central claim rests on experimental verification of robustness rather than any mathematical derivation chain that collapses to its own premises. This is a self-contained algorithmic proposal without the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper introduces no new free parameters or invented entities; it relies on the standard Mapper construction from topological data analysis and the domain assumption that intersection curves can be adequately sampled for topological recovery.

axioms (1)

domain assumption Point samples drawn from the intersection curve of two B-spline surfaces are sufficient to reconstruct its topology via Mapper.
The method depends on this sampling step to feed Mapper; the abstract does not discuss how sampling density or singularity detection is guaranteed.

pith-pipeline@v0.9.0 · 5412 in / 1226 out tokens · 31624 ms · 2026-05-10T14:00:38.352668+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

2 extracted references · 1 canonical work pages

[1]

Intersection of parametric surfaces in the bernstein-bezier representation.Computer-Aided Design, 18(4):186–192, 1986

Dieter Lasser. Intersection of parametric surfaces in the bernstein-bezier representation.Computer-Aided Design, 18(4):186–192, 1986. LuizHenriqueDeFigueiredo. Surfaceintersectionusinganearithmetic. In Proceedingsofgraphicsinterface,volume96,pages168–175.Citeseer, 1996. Hongwei Lin, Yang Qin, Hongwei Liao, and Yunyang Xiong. Affine arithmetic-based b-spli...

1986
[2]

g-mapper: Learning a cover in the mapper construction.SIAM Journal on Mathematics of Data Science, 7(2):572–596, 2025

Palande, Sarah Percival, and Emilie Purvine. g-mapper: Learning a cover in the mapper construction.SIAM Journal on Mathematics of Data Science, 7(2):572–596, 2025. Rachel Jeitziner, Mathieu Carriere, Jacques Rougemont, Steve Oudot, Kathryn Hess, and Cathrin Brisken. Two-tier mapper, an unbiased topology-based clustering method for enhanced global gene exp...

work page arXiv 2025