Visual Complexity of Point Set Mappings

Arjen Simons; Kevin Verbeek; Wouter Meulemans

REVIEW 4 major objections 5 minor 32 references

Visual Complexity of Point Set Mappings

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Counting group moves splits transition problems into easy and hard.

desk verdict A genuinely new problem family for measuring group-motion complexity in point set transitions, with mostly solid algorithmic results and one honest gap: no evidence yet that the measure tracks perception. read the letter →

arxiv 2411.17920 v1 pith:YZK3O3CJ submitted 2024-11-26 cs.CG

Wouter Meulemans , Arjen Simons , Kevin Verbeek This is my paper · ORCID

classification cs.CG MSC 68U0568Q17

keywords grouptranslationsvisualcomplexitypointsetmappingsanimatedtransitionscomputationalgeometryNP-hardnessapproximationalgorithmhierarchicalfamilies

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper argues that the visual complexity of an animated transition between two labeled point sets is better measured by how many groups of points move together than by how far each point travels. It formalizes this as a family of optimization problems: choose a family of subsets (the groups) and a translation per group so that every point's displacement is the sum of the translations of groups containing it, and minimize either the number of groups or the total translation length. The paper maps the algorithmic difficulty of these problems across four structural restrictions on the allowed groups and two cost measures. Most variants are settled with polynomial-time algorithms or NP-hardness proofs; the two remaining open cases receive an approximation algorithm and a hardness result for a restricted version. If the underlying perceptual premise holds, this gives visualization designers a principled way to pick transition strategies that minimize cognitive load.

What carries the argument

The load-bearing object is a solution pair $(\mathcal{F},\tau)$: a family $\mathcal{F}$ of subsets of point indices (the groups) together with a translation vector $\tau(S)$ for each group, subject to the constraint that for every point $i$, the sum of $\tau(S)$ over all groups containing $i$ equals the displacement vector $\delta_i = b_i - a_i$. The cost is either the cardinality $|\mathcal{F}|$ or the total length $\sum_{S\in\mathcal{F}}\|\tau(S)\|$ using the Euclidean norm, with a Manhattan-norm variant used as an approximation tool. The four family constraints — given, disjoint, hierarchical, free — determine the difficulty: disjoint instances collapse to grouping equal vectors, hierarchical instances are represented as rooted trees and solved with interval medians in one dimension, hierarchical instances in two dimensions map onto Steiner trees, and free instances are approximated via coordinate-wise Manhattan solutions. A rank-and-determinant argument transfers $d$-dimensional monotone cardinality instances to one dimension, which is what makes the one-dimensional NP-hardness proof work.

What would settle it

For small $n$, enumerate all possible group families for a random one-dimensional instance and compute the true optimal hierarchical length by brute force; Theorem 9 predicts it equals $\mathrm{span}(\Delta\cup\{0\})$, so any instance where the brute-force optimum is strictly smaller would refute the theorem. Separately, a controlled experiment comparing human ratings of animation simplicity against group-translation cost versus total point-wise distance would test the perceptual premise.

Watch

Extended reading notes

Core claim

The central discovery is that the group-translation model yields a clean computational landscape. For disjoint group families, both minimizing the number of groups and minimizing total length reduce to grouping identical displacement vectors and run in $O(dn\log n)$ time (Theorem 1). For hierarchical families in one dimension, the minimal total length equals $\mathrm{span}(\Delta\cup\{0\})$ and is computable in $O(n\log n)$ time, and the same value is optimal even when the family is unrestricted (Theorem 9). In two dimensions, hierarchical minimal length is NP-hard by an exact equivalence to the Euclidean Steiner tree problem (Theorem 10), while the unrestricted two-dimensional length problem admits a $1.307$-approximation obtained by taking the better of two Manhattan-distance decompositions (Theorem 12). Minimizing the number of groups is polynomial for hierarchical families ($O(dn\log n)$, Theorem 13) but NP-hard for unrestricted families even in one dimension in the monotone case, by a reduction from Vertex Cover (Theorems 14 and 15).

Load-bearing premise

The load-bearing premise is that the minimum number or total length of group translations is a meaningful measure of how complex a transition looks to a human viewer; the algorithmic theorems stand regardless, but the motivating application would collapse if this perceptual link fails.

Editorial extensions

If this is right

In one dimension, both the hierarchical and unrestricted length-minimization problems have optimal value $\mathrm{span}(\Delta\cup\{0\})$, computable in $O(n\log n)$ time.
In two dimensions, hierarchical length minimization is NP-hard because it is exactly the Euclidean Steiner tree problem on the displacement vectors plus the origin.
The unrestricted two-dimensional length problem admits a $1.307$-approximation by solving two Manhattan-distance versions, each in $O(n\log n)$ time.
Minimizing the number of groups is solvable in $O(dn\log n)$ time for disjoint and hierarchical families, but NP-hard for unrestricted families even in one dimension in the monotone case.
The classification extends to new dimensions such as sequential stages, unlabeled point sets, and rotations or scaling, which generate new open problems.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the perceptual premise is accepted, the exact one-dimensional result gives a free design heuristic: in one-dimensional transitions, no group-move animation needs to move points more than the spread of the displacement vectors.
The Steiner-tree equivalence suggests that hard two-dimensional instances are exactly those where intermediate junction points can shorten total movement, so visual complexity in two dimensions has a direct geometric meaning beyond point-wise distance.
A direct testable extension would compare user judgments of animation simplicity against the model's predicted optimal cost for small point sets, isolating whether group count or total length is the better perceptual proxy.
The paper's arc example shows the rotated-Manhattan approximation approach cannot beat $4/\pi$ even with all angles, leaving a narrow gap to the $1.307$ guarantee; a better algorithm would need a different idea.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Tim, quick note on arXiv:2411.17920. The paper is worth sending to a referee, and I would take it as is modulo a few proof patches. The core idea is new: instead of point-set similarity metrics that charge per point, they define transition cost by translating groups of points as single operations, with variants depending on allowed families (given, disjoint, hierarchical, free) and objective (number of groups vs total length). The classification is clean and the algorithmic table is mostly filled.

What is genuinely good: the 1D MLFT/MLHT exact characterization via span(Δ∪{0}) is neat and gives an O(n log n) algorithm; the interval-median O(n) algorithm for 1D hierarchical MLGT is a legitimate extension of median finding; the equivalence between 2D MLHT and Euclidean Steiner tree is exact and makes the NP-hardness honest; the 1.307-approximation for 2D MLFT using two Manhattan orientations is a clever and correct argument. The dimension-reduction trick in Theorem 15 is also a nice technique.

The soft spots are real but not damaging to the algorithmic core. The perceptual premise—that minimum group-translation count or length tracks visual complexity—is asserted, illustrated with a single Common Fate example, and never tested. That caps the applied significance, and the authors should state this limitation as clearly in the introduction as they do implicitly in the body. On the math side: Lemma 4 as printed omits the root translation cost in the lower bound; the proof of Theorem 14 has a sketched edge-reassignment step that needs spelling out; and Theorem 15's statement that the reduced numbers are integers is false (they are rationals with determinant denominators), though the bit-complexity conclusion still holds. None of these break the main results; they are referee-fixable.

Bottom line: this is a solid definitional-algorithmic paper. It belongs in a computational geometry venue despite the missing perceptual validation, because the complexity results stand on their own and the taxonomy opens a real line of work. I would send it out, and ask the authors to fix the proof gaps and to state the validation limitation in the introduction rather than leaving it implicit.

Referee Report

4 major / 5 minor

Summary. The paper proposes a combinatorial framework for measuring the visual complexity of animated transitions between two labeled point sets in R^d. The complexity of a transition is the minimal number or total Euclidean length of translations of groups (subsets) of points that collectively map each starting point to its target. The authors define a taxonomy over family constraints (Given, Disjoint, Hierarchical, Free) and optimization criteria (Cardinality, Length), and they study the resulting algorithmic problems. The results include O(dn log n) algorithms for the Disjoint and Hierarchical Cardinality variants, a convex-optimization formulation for the Given Length variant, an O(n) algorithm for 1D hierarchical Length, an O(n log n) algorithm for 1D Free/Hierarchical Length with optimal cost span(Δ∪{0}), NP-hardness of 2D Hierarchical Length via Euclidean Steiner trees, a 1.307-approximation for 2D Free Length, and NP-hardness of the monotone Free Cardinality problem in arbitrary and then 1D dimension. The paper also discusses two open problems and possible extensions.

Significance. If the proposed measure is accepted as a proxy for the perceptual complexity of animated transitions, this is a valuable and systematic classification that gives a broad algorithmic picture of a natural family of optimization problems. The paper's technical strengths are substantial: the Steiner-tree equivalence in Theorem 10 is exact, the approximation factor in Theorem 12 comes from a clean geometric argument, the 1D length results are tight, and the reductions in Theorems 14 and 15 target a well-known hard problem. The paper contains no fitted parameters or circular reasoning; all results are derived from known mathematics. The main weakness is the unvalidated perceptual premise: the measure is proposed as a model of visual complexity and cognitive load, but no empirical or human-subject evidence supports that connection. This does not invalidate the algorithmic contributions, but it limits the strength of the motivational claim and should be addressed by either evidence or a clear reframing.

major comments (4)

[Section 1, abstract] The manuscript repeatedly asserts that the minimal number or total length of group translations measures 'visual complexity' and can serve as 'a proxy to automatically assess the cognitive load' (Section 1, abstract). No empirical validation is provided; the only supporting evidence is the Gestalt common-fate example in Figure 1. This is load-bearing for the paper's stated application, even though the algorithmic theorems do not depend on the perceptual premise. I recommend adding a user study or, at minimum, substantially softening the claim and explicitly framing the measure as a combinatorial model whose perceptual plausibility is an open hypothesis.
[Section 3, Lemma 4 and Theorem 6] Lemma 4's lower bound omits the cost of the root group translation |τ(S(r))|. As stated, the bound d(x',M(r)) + c cannot justify the optimality of the root translation chosen in Theorem 6. For example, with Δ={10,11} and F={[n],{1},{2}}, the lemma's bound for x'=0 is 11, while the actual cost is 21; for x'=12 the bound is 2, which is below the cost 11 of the algorithm's chosen root translation. The theorem is likely correct, but the proof must minimize |x'| + f(x') rather than f(x') alone, and Lemma 4 should be restated to include the root translation cost.
[Theorem 14, edge reassignment] In the NP-hardness proof for monotone MCFT, the statement 'If an edge has at least 5 groups assigned to it, then we can always reassign one of the groups assigned to e to either u or v instead' is asserted without proof. This is nontrivial because groups that contain edge-specific basis vectors (for example, a group with translation x1(e)+x2(e)) cannot be assigned to u or v, since those vertex displacements do not contain the edge components. Please provide a detailed argument for the reassignment or replace the counting step with a more careful case analysis.
[Theorem 15, full-rank argument] The proof that an optimal solution must have full-rank B states: 'In that case we can add a row to B that corresponds to an equation of the form τ′(S_i)=0.' This is not valid as written: an affine solution space of dimension at least one need not intersect a coordinate hyperplane. The conclusion that a rank-deficient representation can be reduced to fewer groups is plausible and probably true, but the argument should be replaced by a direct column-reduction argument (if the columns are linearly dependent, express Δ' using a basis of the column space).

minor comments (5)

[Section 3, Lemma 3] The notation d(x,M(I)) is used before a distance from a point to an interval is defined; please state explicitly that d(x,[a,b]) extends to intervals by taking the minimum over the interval.
[Theorem 12] The phrase 'of some collection Δ' should read 'of any collection Δ'; otherwise the quantification is ambiguous.
[Section 4, Figure 4] The construction of a Euclidean MLFT of cost approaching π/2 for the two quarter arcs is only sketched; please provide a more explicit description of the groups and translations used in that construction.
[Theorem 9] The sentence 'The resulting family of subsets F may have total size O(n^2), but it can be represented by a tree T with only O(n) nodes' could be clarified: the tree nodes represent the nested chain of sets, and the total size of all sets is O(n^2) only if the sets are listed explicitly.
[Section 5] The monotone MCFT problem is defined informally in the text before Theorem 14; consider adding a formal definition alongside the other problem variants in Section 2.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the algorithmic derivations are self-contained and any perceptual motivation is an explicit assumption, not a derived result.

full rationale

The paper defines a family of optimization problems in which visual complexity is measured by the number or total length of group translations, and then derives algorithmic results about those problems. The derivation chain is self-contained: Theorem 1 follows directly from sorting identical displacement vectors; Theorem 2 observes that the MLGT objective and constraints are convex; Theorems 6 and 9 use interval-median arguments and matching lower and upper bounds; Theorem 10 gives an exact reduction to the Euclidean Steiner tree problem, an independent known NP-hard problem; Theorem 12 derives an approximation factor from norm geometry; and Theorems 14 and 15 reduce from Vertex Cover and construct an algebraic encoding, respectively. No parameter is fitted to data, no prediction is read back from the proposed measure, and no load-bearing premise is justified by a self-citation. The only non-algorithmic premise is that group-translation cost is a plausible proxy for perceived visual complexity, but the paper explicitly proposes this as a modeling choice rather than deriving it, and the algorithmic results do not depend on its empirical validity. Therefore there is no circularity in the mathematical content.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claims rest on a small set of domain assumptions: labeled input, the perceptual validity of group-move cost, and the feasibility convention for given families. No free parameters are fitted to data and no new physical or formal entities are postulated; the group-translation operation is the object of study rather than an unverified input.

assumptions (4)

domain assumption Points are labeled with a fixed correspondence between A and B.
Section 2 states: 'We assume that the points in A and B already have a given mapping such that a_i should be translated to b_i.' This reduces the input to a set of displacement vectors and removes correspondence as an algorithmic issue.
domain assumption The cost of moving a group of points together is a valid proxy for visual complexity.
Section 1 motivates the model with the Gestalt principle of Common Fate (Figure 1) but provides no user study or perceptual validation; the entire application framing rests on this assumption.
domain assumption For the Given variant, F contains all singleton sets to guarantee feasibility.
Section 2 states: 'To ensure feasibility of the problem, we will assume that F always contains at least all singleton sets.' This is a modeling convention rather than an intrinsic property of the application.
standard math NP-hardness of Euclidean and rectilinear Steiner tree and of Vertex Cover is taken from the literature.
Theorems 10 and 14 reduce from known NP-hard problems; correctness relies on those hardness results.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Visual Complexity of Point Set Mappings." pith.science (2026). https://pith.science/paper/YZK3O3CJ

@misc{pith2026241117920,
  author       = {Pith},
  title        = {Pith review of: Visual Complexity of Point Set Mappings},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YZK3O3CJ}},
  note         = {Machine review of arXiv:2411.17920}
}

read the original abstract

We study the visual complexity of animated transitions between point sets. Although there exist many metrics for point set similarity, these metrics are not adequate for this purpose, as they typically treat each point separately. Instead, we propose to look at translations of entire subsets/groups of points to measure the visual complexity of a transition between two point sets. Specifically, given two labeled point sets A and B in R^d, the goal is to compute the cheapest transformation that maps all points in A to their corresponding point in B, where the translation of a group of points counts as a single operation in terms of complexity. In this paper we identify several problem dimensions involving group translations that may be relevant to various applications, and study the algorithmic complexity of the resulting problems. Specifically, we consider different restrictions on the groups that can be translated, and different optimization functions. For most of the resulting problem variants we are able to provide polynomial time algorithms, or establish that they are NP-hard. For the remaining open problems we either provide an approximation algorithm or establish the NP-hardness of a restricted version of the problem. Furthermore, our problem classification can easily be extended with additional problem dimensions giving rise to new problem variants that can be studied in future work.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

[1]

In: Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques

Alexa, M., Cohen-Or, D., Levin, D.: As-rigid-as-possible shape interpolation. In: Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques. p. 157–164. SIGGRAPH '00, ACM Press/Addison-Wesley Publishing Co., USA (2000)

work page 2000
[2]

Annals of Mathematics and Artificial Intelligence 13, 251--265 (1995)

Alt, H., Behrends, B., Blömer, J.: A pproximate matching of polygonal shapes. Annals of Mathematics and Artificial Intelligence 13, 251--265 (1995)

work page 1995
[3]

In: Handbook of computational geometry, pp

Alt, H., Guibas, L.J.: Discrete geometric shapes: Matching, interpolation, and approximation. In: Handbook of computational geometry, pp. 121--153. Elsevier (2000)

work page 2000
[4]

In: International Symposium on Graph Drawing

Archambault, D., Purchase, H.C.: Mental map preservation helps user orientation in dynamic graphs. In: International Symposium on Graph Drawing. pp. 475--486. Springer, Berlin, Heidelberg (2013)

work page 2013
[5]

IEEE Transactions on Pattern Analysis and Machine Intelligence PAMI-9(5), 698--700 (1987)

Arun, K.S., Huang, T.S., Blostein, S.D.: Least-squares fitting of two 3- D point sets. IEEE Transactions on Pattern Analysis and Machine Intelligence PAMI-9(5), 698--700 (1987)

work page 1987
[6]

IEEE Transactions on Pattern Analysis and Machine Intelligence 14(2), 239--256 (1992)

Besl, P.J., McKay, N.D.: A method for registration of 3- D shapes. IEEE Transactions on Pattern Analysis and Machine Intelligence 14(2), 239--256 (1992)

work page 1992
[7]

Operations Research 29 (1981)

Bland, R., Goldfarb, D., Todd, M.: Feature article—the ellipsoid method: a survey. Operations Research 29 (1981)

work page 1981
[8]

In: Proc

Cohen, S., Guibas, L.: The E arth M over's distance under transformation sets. In: Proc. 7th IEEE International Conference on Computer Vision. vol. 2, pp. 1076--1083. IEEE (1999)

work page 1999

Show all 32 references

[9]

In: Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems

Du, F., Cao, N., Zhao, J., Lin, Y.R.: Trajectory bundling for animated transitions. In: Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems. p. 289–298. CHI '15, Association for Computing Machinery, New York, NY, USA (2015)

2015
[10]

In: Proceedings of the 9th International Conference on Geographic Information Science

Duckham, M., van Kreveld, M.J., Purves, R., Speckmann, B., Tao, Y., Verbeek, K., Wood, J.: Modeling checkpoint-based movement with the E arth M over's D istance. In: Proceedings of the 9th International Conference on Geographic Information Science. pp. 225--239. LNCS 9927, Spr...

2016
[11]

International Journal of Computational Geometry & Applications 25(02), 101--122 (2015)

Eppstein, D., van Kreveld, M., Speckmann, B., Staals, F.: Improved grid map layout by point set matching. International Journal of Computational Geometry & Applications 25(02), 101--122 (2015)

2015
[12]

Psychonomic Bulletin and Review 15, 802--808 (2008)

Franconeri, S., Lin, J., Pylyshyn, Z., Fisher, B., Enns, J.: Evidence against a speed limit in multiple-object tracking. Psychonomic Bulletin and Review 15, 802--808 (2008)

2008
[13]

SIAM Journal on Applied Mathematics 32(4), 835--859 (1977)

Garey, M.R., Graham, R.L., Johnson, D.S.: The complexity of computing S teiner minimal trees. SIAM Journal on Applied Mathematics 32(4), 835--859 (1977)

1977
[14]

SIAM Journal on Applied Mathematics 32(4), 826--834 (1977)

Garey, M.R., Johnson, D.S.: The rectilinear S teiner tree problem is NP -complete. SIAM Journal on Applied Mathematics 32(4), 826--834 (1977)

1977
[15]

Computers & Graphics 25(1), 67--75 (2001)

Gotsman, C., Surazhsky, V.: Guaranteed intersection-free polygon morphing. Computers & Graphics 25(1), 67--75 (2001)

2001
[16]

IEEE Transactions on Visualization and Computer Graphics 13(6), 1240--1247 (2007)

Heer, J., Robertson, G.: Animated transitions in statistical data graphics. IEEE Transactions on Visualization and Computer Graphics 13(6), 1240--1247 (2007)

2007
[17]

Journal of the Optical Society A 4, 629--642 (1987)

Horn, B.: Closed-form solution of absolute orientation using unit quaternions. Journal of the Optical Society A 4, 629--642 (1987)

1987
[18]

IEEE Transactions on Pattern Analysis and Machine Intelligence 33(8), 1633--1645 (2011)

Jian, B., Vemuri, B.C.: Robust point set registration using G aussian mixture models. IEEE Transactions on Pattern Analysis and Machine Intelligence 33(8), 1633--1645 (2011)

2011
[19]

SIAM Journal on Computing 22(6), 1117--1141 (1993)

Jiang, T., Ravikumar, B.: Minimal NFA problems are hard. SIAM Journal on Computing 22(6), 1117--1141 (1993)

1993
[20]

In: Miller, R.E., Thatcher, J.W., Bohlinger, J.D

Karp, R.M.: Reducibility among combinatorial problems. In: Miller, R.E., Thatcher, J.W., Bohlinger, J.D. (eds.) Complexity of Computer Computations. The IBM Research Symposia Series. pp. 85--103. Springer, Boston, MA (1972)

1972
[21]

IEEE Transactions on Systems, Man, and Cybernetics: Systems 54(7), 4101--4112 (2024)

Li, L., Yang, M., Wang, C.: Graph correspondence-based point set registration. IEEE Transactions on Systems, Man, and Cybernetics: Systems 54(7), 4101--4112 (2024)

2024
[22]

Computers & Graphics 76, 60--72 (2018)

Liu, Z., Zhou, L., Leung, H., Shum, H.P.: High-quality compatible triangulations and their application in interactive animation. Computers & Graphics 76, 60--72 (2018)

2018
[23]

Computer Graphics Forum 38, 713--723 (2019)

Meulemans, W.: Efficient optimal overlap removal: Algorithms and experiments. Computer Graphics Forum 38, 713--723 (2019)

2019
[24]

IEEE Transactions on Visualization and Computer Graphics 23(1), 381--390 (2017)

Meulemans, W., Dykes, J., Slingsby, A., Turkay, C., Wood, J.: Small multiples with gaps. IEEE Transactions on Visualization and Computer Graphics 23(1), 381--390 (2017)

2017
[25]

In: Computer Graphics Forum

Mizuno, K., Wu, H.Y., Takahashi, S., Igarashi, T.: Optimizing stepwise animation in dynamic set diagrams. In: Computer Graphics Forum. vol. 38, pp. 13--24. Wiley Online Library (2019)

2019
[26]

AK Peters/CRC Press (2014)

Munzner, T.: Visualization analysis and design. AK Peters/CRC Press (2014)

2014
[27]

IEEE Transactions on Pattern Analysis and Machine Intelligence 32(12), 2262--2275 (2010)

Myronenko, A., Song, X.: Point set registration: Coherent point drift. IEEE Transactions on Pattern Analysis and Machine Intelligence 32(12), 2262--2275 (2010)

2010
[28]

International Journal of Computer Vision 40, 99–121 (2000)

Rubner, Y., Tomasi, C., Guibas, L.: The E arth M over’s D istance as a metric for image retrieval. International Journal of Computer Vision 40, 99–121 (2000)

2000
[29]

Master's thesis, TU Eindhoven (2020)

Suiker, K.N.: Optimizing Staged Transition for Scatter Plots. Master's thesis, TU Eindhoven (2020)

2020
[30]

In: Spatial Information Theory: A Theoretical Basis for GIS

Tversky, B.: Cognitive maps, cognitive collages, and spatial mental models. In: Spatial Information Theory: A Theoretical Basis for GIS. pp. 14--24. LNCS 716, Springer, Berlin, Heidelberg (1993)

1993
[31]

IEEE Transactions on Visualization and Computer Graphics 24(9), 2487--2500 (2018)

Wang, Y., Archambault, D., Scheidegger, C.E., Qu, H.: A vector field design approach to animated transitions. IEEE Transactions on Visualization and Computer Graphics 24(9), 2487--2500 (2018)

2018
[32]

Palm and Enke, Erlangen, Germany (1925)

Wertheimer, M.: Drei Abhandlungen zur Gestalttheorie. Palm and Enke, Erlangen, Germany (1925)

1925

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

In: Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques

Alexa, M., Cohen-Or, D., Levin, D.: As-rigid-as-possible shape interpolation. In: Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques. p. 157–164. SIGGRAPH '00, ACM Press/Addison-Wesley Publishing Co., USA (2000)

work page 2000

[2] [2]

Annals of Mathematics and Artificial Intelligence 13, 251--265 (1995)

Alt, H., Behrends, B., Blömer, J.: A pproximate matching of polygonal shapes. Annals of Mathematics and Artificial Intelligence 13, 251--265 (1995)

work page 1995

[3] [3]

In: Handbook of computational geometry, pp

Alt, H., Guibas, L.J.: Discrete geometric shapes: Matching, interpolation, and approximation. In: Handbook of computational geometry, pp. 121--153. Elsevier (2000)

work page 2000

[4] [4]

In: International Symposium on Graph Drawing

Archambault, D., Purchase, H.C.: Mental map preservation helps user orientation in dynamic graphs. In: International Symposium on Graph Drawing. pp. 475--486. Springer, Berlin, Heidelberg (2013)

work page 2013

[5] [5]

IEEE Transactions on Pattern Analysis and Machine Intelligence PAMI-9(5), 698--700 (1987)

Arun, K.S., Huang, T.S., Blostein, S.D.: Least-squares fitting of two 3- D point sets. IEEE Transactions on Pattern Analysis and Machine Intelligence PAMI-9(5), 698--700 (1987)

work page 1987

[6] [6]

IEEE Transactions on Pattern Analysis and Machine Intelligence 14(2), 239--256 (1992)

Besl, P.J., McKay, N.D.: A method for registration of 3- D shapes. IEEE Transactions on Pattern Analysis and Machine Intelligence 14(2), 239--256 (1992)

work page 1992

[7] [7]

Operations Research 29 (1981)

Bland, R., Goldfarb, D., Todd, M.: Feature article—the ellipsoid method: a survey. Operations Research 29 (1981)

work page 1981

[8] [8]

In: Proc

Cohen, S., Guibas, L.: The E arth M over's distance under transformation sets. In: Proc. 7th IEEE International Conference on Computer Vision. vol. 2, pp. 1076--1083. IEEE (1999)

work page 1999