Pith. sign in

REVIEW 4 major objections 6 minor 35 references

Swept Volume Computation with Enhanced Geometric Detail Preservation

T0 review · 4 major / 6 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read A point lies in the swept volume exactly when its backward-in-time path hits the stationary object, and the paper builds a practical algorithm on that identity.

desk verdict A promising multi-field extraction idea with an unquantified linearization step and an appraisal that doesn't yet support the headline robustness claims. read the letter →

arxiv 2509.09325 v1 pith:3EPULHKE submitted 2025-09-11 cs.CG

Pengfei Wang , Yuexin Yang , Shuangmin Chen , Shiqing Xin , Changhe Tu , Wenping Wang This is my paper
classification cs.CG MSC 68U0565D18
keywords sweptvolumeinversemotiondistancefieldisosurfaceextractiontetrahedralmesh4Dincrementalcuttingrigidcollisiondetection
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

The paper proposes computing swept volumes by inverting the motion: keep the object still and let each spatial point travel backward along the inverse trajectory. A point belongs to the swept volume exactly when its backward path intersects the static object, and over a short time interval that path can be treated as a straight segment. The distance from a point to the swept volume is then approximated by the minimum distance from that segment to the object. To preserve fine geometric details at trajectory intersections, the method stores multiple signed distance fields per tetrahedron, combining them with a pointwise minimum and extracting the zero surface by incremental cutting in 4D. If correct, robust swept-volume computation reduces to segment-to-model distance queries plus a multi-field isosurface extraction, with direct applications in robot path planning and collision detection.

What carries the argument

Motion perspective inversion: the exact statement that swept-volume membership equals intersection of the inverse point trajectory with the static model. The method then uses piecewise-linear inverse trajectories, a multi-field tetrahedral representation where each temporal segment contributes its own linear distance field inside a tetrahedron, the pointwise minimum of those fields as the combined distance, and a 4D incremental cutting procedure that builds the lower envelope of hyperplanes and intersects it with the zero hyperplane to extract the boundary surface.

What would settle it

Take a rigid motion with high angular velocity—for instance, a sphere rotating quickly around an off-center axis—and compute the swept volume with a single time step using the linear-segment approximation. Then compare the exact inverse trajectory membership for points near the boundary against the linear-segment prediction: if any point is classified inside when its true curved inverse path never touches the model, the approximation is demonstrably invalid at that resolution.

Watch

Extended reading notes

Core claim

The central discovery is an exact membership criterion for rigid motion: for a transformation f(t), a point q is inside the swept volume over a time interval exactly when its inverse trajectory q(t)=f^{-1}(t)q crosses the static model M. Approximating each inverse trajectory by one line segment per short subinterval turns distance-to-swept-volume into a segment-to-model distance, and the complete swept volume is built by taking the minimum of the resulting segment-wise distance fields. The paper argues that this inversion, combined with maintaining several such fields inside each tetrahedron and extracting the surface via 4D incremental cutting, recovers sharp features and smooth boundaries

Load-bearing premise

The load-bearing premise is that every inverse trajectory over a chosen time interval is well approximated by a straight segment; the paper gives no bound connecting time-step size to motion curvature or model geometry, and its own ablation shows that coarse time steps produce large deviations.

Editorial extensions

If this is right

  • Swept-volume computation becomes a pipeline of standard geometric primitives: segment-to-model distance queries followed by local isosurface extraction.
  • Multiple distance fields per tetrahedron allow sharp features and creases that appear where different temporal segments meet to survive in the extracted mesh.
  • The method works with mesh input and mesh output, which makes the result directly usable in downstream path planning and collision detection systems.
  • Extraction is independent per tetrahedron, so the surface reconstruction step parallelizes across cores.
  • Reported experiments, including a sphere on a circular trajectory with known analytical swept volume, show lower Chamfer and Hausdorff errors and shorter compute times than the two compared baselines.

Reading between the lines

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

  • If the linear-trajectory approximation is the crux, then adaptive temporal refinement driven by local motion curvature or speed should cut the main error source; the paper identifies uniform temporal discretization as a limitation but does not test adaptation.
  • The exact inverse-trajectory membership criterion could be paired with a spatial acceleration structure on the static model to answer point-membership queries interactively, which the paper does not explore.
  • The same multi-field minimum-combination idea may extend to unions of multiple swept segments, Minkowski sums, or other constructive solid geometry operations beyond the tested cases.
  • A formal error bound relating time-step size, motion curvature, and model feature size would turn the heuristic approximation into a certified algorithm; no such bound is provided.
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.

Referee Report

4 major / 6 minor

Summary. The paper proposes a swept volume computation method based on motion perspective inversion. Instead of tracking the moving object, the method fixes the object and traces spatial query points backward in time. The inverse trajectory is temporally subdivided and approximated by linear segments; for each segment, the signed distance from a query point to the swept volume is approximated by the minimum signed distance from the segment to the static model (Eq. 3). The method then maintains multiple distance fields per tetrahedron, one per temporal segment, and extracts the zero isosurface via 4D incremental cutting. The authors claim the method robustly handles diverse motions, including translations and screw motions, while preserving fine geometric details at trajectory intersections, and they validate with visual comparisons, one quantitative analytical case, and ablation studies.

Significance. If the central approximation and multi-field extraction are correct, the perspective inversion is an attractive alternative to explicit surface tracking and global optimization-based SDF methods. The multi-field representation is a sensible way to capture features created by different temporal portions of the motion, and the 4D incremental cutting approach is a natural extension of the authors' prior work. The paper is honest about its limitations (uniform discretization, inability to handle deformation). However, the significance is currently limited by the lack of error bounds for the central distance approximation, the absence of quantitative validation for non-translational motions, and the reliance on a single analytical test case for numerical claims.

major comments (4)
  1. [§4.1, Eq. (3)] The central approximation d(q, SweepVol(M,[t0,t1])) ≈ min_{p in L} d(p,M) is exact in sign but not in magnitude for general rigid motions. For overlapping component swept volumes, the signed distance to a union is not the minimum of the component signed distances: e.g. for intervals [0,2] and [1,3], the min of the two SDFs at q=1.5 is -0.5, while the true signed distance to the union [0,3] is -1.5. Since Algorithm 1 interpolates vertex distance values linearly to locate zero crossings, incorrect magnitudes can shift the extracted surface. No bound is given linking the temporal segment length, motion curvature/angular velocity, or model geometry to this error. The paper must either prove that the zero set is independent of the magnitude error, or provide an error bound and quantitative tests for rotations/screw motions.
  2. [§5.2, Algorithm 2] When the line segment partially lies inside the model, the algorithm samples 10 evenly distributed points on the interior portion and takes the minimum signed distance at these samples. This introduces an uncontrolled error: the true minimum can occur between samples, and the signed distance to the model is not linear along the segment. The paper states that the sign remains correct, but the magnitude error directly affects the linear interpolation used for zero-crossing extraction in §4.2. An error estimate in terms of the model's local curvature and the segment length is needed to support the claimed fidelity.
  3. [§6.2, Table 1] The only quantitative benchmark is a sphere moving along a circular trajectory. Since a sphere is rotationally symmetric, this is effectively a translation and does not exercise the rotational/screw-motion error of Eq. (3). The visual comparisons in Figure 9 are restricted to translations, while the abstract claims screw motions. Quantitative comparisons with [SAJ21] and Stamping for screw motions are missing. Without such validation, the claim of detail preservation for screw motions rests only on visual examples.
  4. [§6.4, Fig. 13] The ablation study for temporal discretization is visual only. It shows convergence behavior but gives no quantitative error as a function of N. Given that N is a free parameter and the central approximation error is uncontrolled, the paper should report a numerical error curve (e.g., Chamfer or Hausdorff distance vs. N) for at least one non-translational motion.
minor comments (6)
  1. [Abstract] Typo: 'distance fileds' should be 'distance fields'.
  2. [Eq. (2)] The notation L(t0,t1)=q(t0)q(t1) is ambiguous; use an overline or explicit segment notation.
  3. [Fig. 7 caption] 'can be into five tetrahedra' should be 'can be subdivided into five tetrahedra'.
  4. [§5.2, Algorithm 2] L_interior is not formally defined; specify that it is the subset of L lying inside M.
  5. [§6.2] The sentence 'we limit the motion to translation to maintain trajectory consistency' should be reconciled with the paper's broader claim of handling screw motions; clarify why the comparison is restricted.
  6. [Table 1] Units are given inconsistently: CD in ‰ and HD in %. Use the same normalization convention for both, or explain the difference.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central distance computation is geometrically derived and checked against an analytic swept surface; the self-citations are implementation reuse, not load-bearing for the claimed result.

full rationale

The core claim (Section 4.1) that q lies in the swept volume iff its inverse trajectory intersects the static model follows from rigid-motion invertibility (Eq. 1), not from any fitted parameter or prior result. Equation 3 is explicitly an approximation using the chord of the inverse trajectory, and Algorithm 2 computes segment-to-model distances with FCPW plus signed-distance sampling; no quantity is fit to ground-truth swept volumes and then re-reported as a prediction. The quantitative evaluation uses an analytically defined sphere-on-circular-path surface, which provides an external reference. The citations to [WSW*25] and [XWX*22] are confined to the linear-interpolation assumption, the competition culling rule, and the 4D incremental-cutting extraction; adopting a numerical extraction routine from prior work does not make the swept-volume distance computation self-referential. The main weakness is an unquantified linearization error for curved/rotational motions and an ablation that is visual rather than metric; that is a correctness/accuracy concern, not circularity.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The method introduces no new physical entities. Its central claim rests on the motion-inversion equivalence, the piecewise-linear approximation of inverse trajectories, and the linear-field interpolation, plus algorithmic components taken from the authors' prior work. The free parameters are discretization and sampling choices that affect the reported accuracy.

free parameters (4)
  • temporal_discretization_N = 50 in main experiments
    The number of time segments into which the motion is divided. Approximation accuracy depends on it; ablation shows changing N changes the result. It is a user-chosen resolution, not fitted to data, but reported metrics use N=50.
  • spatial_resolution = 256^3
    Grid resolution for tetrahedral tessellation. Results vary with this parameter (Figure 14). Chosen as the highest in the ablation.
  • seed_tetrahedra_count = 100
    Fixed number of random seed tetrahedra per time interval in Algorithm 3. The paper does not ablate this choice, and it likely affects coverage and final result.
  • interior_sample_points = 10
    Number of points sampled along the line segment portion inside the model in Algorithm 2. The min signed distance is approximated by the minimum over these samples; increasing the count would improve accuracy at the cost of speed.
assumptions (5)
  • standard math A point lies inside the swept volume iff its inverse trajectory intersects the static model.
    Section 4.1: stated as 'a critical observation'. This is a direct consequence of the definition of swept volume and the relativity of motion.
  • domain assumption Inverse trajectories can be approximated as linear segments over short time intervals.
    Section 4.1 and Section 5.1. All distance computations rely on this approximation; no error bound is provided.
  • domain assumption Each distance field varies linearly within a tetrahedron (Eq. 4).
    Section 4.2, following the linear interpolation assumption in the authors' prior work. This is an approximation that affects the 4D cutting accuracy.
  • standard math The 4D incremental cutting procedure correctly extracts the zero-isosurface of the lower envelope of hyperplanes.
    Section 4.2 and Algorithm 1, following [WSW*25]. The paper does not prove this here; it relies on the cited paper.
  • domain assumption FCPW library returns accurate segment-to-model distances for the outside case.
    Section 5.2, Algorithm 2. The computation uses an external library without validating its accuracy for all inputs.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Swept Volume Computation with Enhanced Geometric Detail Preservation." pith.science (2026). https://pith.science/paper/3EPULHKE

@misc{pith2026250909325,
  author       = {Pith},
  title        = {Pith review of: Swept Volume Computation with Enhanced Geometric Detail Preservation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3EPULHKE}},
  note         = {Machine review of arXiv:2509.09325}
}
read the original abstract

Swept volume computation, the determination of regions occupied by moving objects, is essential in graphics, robotics, and manufacturing. Existing approaches either explicitly track surfaces, suffering from robustness issues under complex interactions, or employ implicit representations that trade off geometric fidelity and face optimization difficulties. We propose a novel inversion of motion perspective: rather than tracking object motion, we fix the object and trace spatial points backward in time, reducing complex trajectories to efficiently linearizable point motions. Based on this, we introduce a multi field tetrahedral framework that maintains multiple distance fileds per element, preserving fine geometric details at trajectory intersections where single field methods fail. Our method robustly computes swept volumes for diverse motions, including translations and screw motions, and enables practical applications in path planning and collision detection.

Figures

Figures reproduced from arXiv: 2509.09325 by the authors.

Figure 1
Figure 1. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Real-world examples: (a) robot vacuum navigation; (b) autonomous vehicle parking. face extraction strategies applied to the resulting SDFs struggle to preserve complex geometric details. In this work, drawing inspiration from inverse analysis ap￾proaches in ray tracing methods [Bar86] and existing swept volume computation techniques [SAJ21], and addressing the limitations of single-field extraction methods, we propo… view at source ↗
Figure 3
Figure 3. Distance Computation. (a) Given a model (purple circle as an example) and its sweeping trajectory, if we could compute the distance from any spatial point (exemplified by the red point) to the resulting swept volume, we would be able to construct the distance field. However, this is extremely challenging without the explicit swept volume result. (b) Exploiting the relativity of motion, we invert the motion perspecti… view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Illustration of our relative motion analysis for short time intervals. (a) Generate the swept volume by moving the object along the specified trajectory. (b) Keep the object stationary and move the observation point p along the reverse trajectory. (c) Ap￾proximate the …
Figure 5
Figure 5. Figure 5: Multiple distance field representation in our approach. (a) Input model and motion trajectory. (b) Complete swept volume. (c) The entire motion is temporally partitioned into multiple seg￾ments, with each segment generating a partial swept volume. Each partial swept vo…
Figure 6
Figure 6. Figure 6: Feature preservation through multiple distance fields and incremental cutting in 2D swept volume extraction. Taking the highlighted triangle as an example, it stores two distance fields from swept volumes during different time segments, with trajectory segments from di…
Figure 8
Figure 8. Figure 8: Temporal discretization. The green model remains sta￾tionary while orange curves show inverse trajectories of spatial points. Within each discrete time interval, these trajectories are ap￾proximated as linear segments. Algorithm 2 Line Segment to Model Distance Query R…
Figure 9
Figure 9. Figure 9: Visual comparison of swept volumes generated by our method against Stamping and [SAJ21] across different models and motion types. Our method preserves sharp features while avoiding the zigzag artifacts present in the Stamping approach. Algorithm 3 Distance Field Propag…
Figure 10
Figure 10. Figure 10: Diverse swept volume results generated by our algorithm for various geometric models undergoing different motion trajectories [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: Swept volume generated by a sphere traversing a circu￾lar trajectory. This configuration yields an analytically defined sur￾face, facilitating quantitative evaluation through direct point sam￾pling without explicit construction of a reference model [PITH_FULL_IMAGE:f…
Figure 14
Figure 14. Figure 14: Extracted sweep volume at different spatial resolutions with temporal resolution fixed at 50. Flat rendering is used to better highlight the differences across resolutions [PITH_FULL_IMAGE:figures/full_fig_p010_14.png]
Figure 15
Figure 15. Figure 15: Visual comparison across different temporal and spatial discretization resolutions. Flat rendering is used to better highlight the differences across resolutions [PITH_FULL_IMAGE:figures/full_fig_p010_15.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

35 extracted references · 4 canonical work pages

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry add.period write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.state := if if FUNCTION not #0 #1 if FUNCTION and 'skip pop #0 if FUNCTIO...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry.original add.period write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.state := if if FUNCTION not #0 #1 if FUNCTION and 'skip pop #0 i...

  3. [3]

    Abrams S., Allen P. K. : Computing swept volumes. The Journal of Visualization and Computer Animation 11, 2 (2000), 69--82

    2000

  4. [4]

    : A computational framework for boundary representation of solid sweeps

    Adsul B., Machchhar J., Sohoni M. : A computational framework for boundary representation of solid sweeps. Computer-Aided Design and Applications 12 (03 2014). https://doi.org/10.1080/16864360.2014.962430 doi:10.1080/16864360.2014.962430

    arXiv 2014

  5. [5]

    : Path trajectory verification for robot manipulators in a manufacturing environment

    Abdel-Malek K., Yeh H. : Path trajectory verification for robot manipulators in a manufacturing environment. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 211, 7 (1997), 547--556

    1997

  6. [6]

    Abdel-Malek K., Yang J., Blackmore D. L. : On swept volume formulations: implicit surfaces. Comput. Aided Des. 33 (2001), 113--121

    2001

  7. [7]

    L., Joy K

    Abdel-Malek K., Yang J., Blackmore D. L., Joy K. I. : Swept volumes: Fundation, perspectives, and applications. Int. J. Shape Model. 12 (2006), 87--127

    2006

  8. [8]

    Barr A. H. : Ray tracing deformed surfaces. In Proceedings of the 13th Annual Conference on Computer Graphics and Interactive Techniques (New York, NY, USA, 1986), SIGGRAPH '86, Association for Computing Machinery, p. 287–296. URL: https://doi.org/10.1145/15922.15918, https://doi.org/10.1145/15922.15918 doi:10.1145/15922.15918

    arXiv 1986

Show all 35 references
  1. [9]

    : Polygonal boundary evaluation of minkowski sums and swept volumes

    Campen M., Kobbelt L. : Polygonal boundary evaluation of minkowski sums and swept volumes. Computer Graphics Forum 29, 5 (2010), 1613--1622. https://doi.org/https://doi.org/10.1111/j.1467-8659.2010.01770.x doi:https://doi.org/10.1111/j.1467-8659.2010.01770.x

    2010

  2. [10]

    R., Lopes D

    de Ara\' u jo B. R., Lopes D. S., Jepp P., Jorge J. A., Wyvill B. : A survey on implicit surface polygonization. ACM Comput. Surv. 47, 4 (May 2015). URL: https://doi.org/10.1145/2732197, https://doi.org/10.1145/2732197 doi:10.1145/2732197

    2015 doi

  3. [11]

    : Online motion planning based on swept volume search with replanning using sequential quadratic programming

    Iwasaki T., Takase Y., Arnold S., Takeshita K., Yamazaki K. : Online motion planning based on swept volume search with replanning using sequential quadratic programming. Advanced Robotics 37 (2023), 737 -- 750

    2023

  4. [12]

    : Dual contouring of hermite data

    Ju T., Losasso F., Schaefer S., Warren J. : Dual contouring of hermite data. ACM Trans. Graph. 21, 3 (July 2002), 339–346. URL: https://doi.org/10.1145/566654.566586, https://doi.org/10.1145/566654.566586 doi:10.1145/566654.566586

    2002

  5. [13]

    : Neural implicit swept volume models for fast collision detection

    Joho D., Schwinn J., Safronov K. : Neural implicit swept volume models for fast collision detection. 2024 IEEE International Conference on Robotics and Automation (ICRA) (2024), 15402--15408. URL: https://api.semanticscholar.org/CorpusID:267897342

    2024

  6. [14]

    : Computing Minkowski sums

    Kaul A. : Computing Minkowski sums. PhD thesis, USA, 1993. UMI Order No. GAX93-33799

    1993

  7. [15]

    Korein J. U. : A geometric investigation of reach. MIT Press, Cambridge, MA, USA, 1986

    1986

  8. [16]

    J., Redon S., Lin M

    Kim Y. J., Redon S., Lin M. C., Manocha D., Templeman J. N. : Interactive continuous collision detection using swept volume for avatars. PRESENCE: Teleoperators and Virtual Environments 16 (2007), 206--223

    2007

  9. [17]

    : An implicit representation of swept volumes based on local shapes and movements

    Laroche C. : An implicit representation of swept volumes based on local shapes and movements. ArXiv abs/2003.11527 (2020)

    2003 arXiv

  10. [18]

    E., Cline H

    Lorensen W. E., Cline H. E. : Marching cubes: A high resolution 3d surface construction algorithm. In Proceedings of the 14th Annual Conference on Computer Graphics and Interactive Techniques (New York, NY, USA, 1987), SIGGRAPH '87, Association for Computing Machinery, p. 163–...

    1987

  11. [19]

    J., Konidaris G

    Murray S., Floyd-Jones W., Qi Y., Sorin D. J., Konidaris G. D. : Robot motion planning on a chip. In Robotics: Science and Systems (2016)

    2016

  12. [20]

    J., Konidaris G

    Murray S., Floyd-Jones W., Qi Y., Sorin D. J., Konidaris G. D. : Robot motion planning on a chip. In Robotics: Science and Systems (2016), vol. 6

    2016

  13. [21]

    : Interactive swept surface modeling in virtual reality with motion-tracked controllers

    McGraw T., Garcia E., Sumner D. : Interactive swept surface modeling in virtual reality with motion-tracked controllers. In Proceedings of the Symposium on Sketch-Based Interfaces and Modeling (2017), pp. 1--9

    2017

  14. [22]

    : Swept volumes

    Peternell M., Pottmann H., Steiner T., Zhao H. : Swept volumes. Computer-Aided Design & Applications 2 (01 2005). https://doi.org/10.1080/16864360.2005.10738324 doi:10.1080/16864360.2005.10738324

    2005

  15. [23]

    : Swept volumes via spacetime numerical continuation

    Sell\' a n S., Aigerman N., Jacobson A. : Swept volumes via spacetime numerical continuation. ACM Trans. Graph. 40, 4 (July 2021). https://doi.org/10.1145/3450626.3459780 doi:10.1145/3450626.3459780

    2021

  16. [24]

    : Fcpw: Fastest closest points in the west, 2021

    Sawhney R. : Fcpw: Fastest closest points in the west, 2021

    2021

  17. [25]

    Y., Gaddipati V

    Shih F. Y., Gaddipati V. : Geometric modeling and representation based on sweep mathematical morphology. Information Sciences 171, 1 (2005), 213--231. https://doi.org/https://doi.org/10.1016/j.ins.2004.04.002 doi:https://doi.org/10.1016/j.ins.2004.04.002

    2005 doi

  18. [26]

    : Implicit modeling of swept surfaces and volumes

    Schroeder W., Lorensen W., Linthicum S. : Implicit modeling of swept surfaces and volumes. In Proceedings Visualization '94 (1994), pp. 40--45. https://doi.org/10.1109/VISUAL.1994.346339 doi:10.1109/VISUAL.1994.346339

    1994

  19. [27]

    : Function representation for sweeping by a moving solid

    Sourin A., Pasko A. : Function representation for sweeping by a moving solid. IEEE Transactions on Visualization and Computer Graphics 2, 1 (1996), 11--18. https://doi.org/10.1109/2945.489382 doi:10.1109/2945.489382

    1996

  20. [28]

    : High quality conservative surface mesh generation for swept volumes

    von Dziegielewski A., Hemmer M., Schömer E. : High quality conservative surface mesh generation for swept volumes. In 2012 IEEE International Conference on Robotics and Automation (2012), pp. 764--769. https://doi.org/10.1109/ICRA.2012.6224921 doi:10.1109/ICRA.2012.6224921

    2012

  21. [29]

    : The differential equation algorithm for general deformed swept volumes

    Wang G., Hua X., Sun J. : The differential equation algorithm for general deformed swept volumes. Journal of Computer Science and Technology 15 (2000), 604--610

    2000

  22. [30]

    D., Leu M

    Weld J. D., Leu M. C. : Geometric representation of swept volumes with application to polyhedral objects. Int. J. Rob. Res. 9, 5 (Sept. 1990), 105–117. https://doi.org/10.1177/027836499000900507 doi:10.1177/027836499000900507

    1990 doi

  23. [31]

    : Data structure for soft objects

    Wyvill G., McPheeters C., Wyvill B. : Data structure for soft objects. The Visual Computer - VC 2 (08 1986), 227--234. https://doi.org/10.1007/BF01900346 doi:10.1007/BF01900346

    1986 doi

  24. [32]

    : Towards voronoi diagrams of surface patches

    Wang P., Song J., Wang L., Xin S., Yan D.-M., Chen S., Tu C., Wang W. : Towards voronoi diagrams of surface patches. IEEE Transactions on Visualization and Computer Graphics (2025), 1--15. https://doi.org/10.1109/TVCG.2025.3531445 doi:10.1109/TVCG.2025.3531445

    2025

  25. [33]

    : Implicit swept volume sdf: Enabling continuous collision-free trajectory generation for arbitrary shapes

    Wang J., Zhang T., Zhang Q., Zeng C., Yu J., Xu C., Xu L., Gao F. : Implicit swept volume sdf: Enabling continuous collision-free trajectory generation for arbitrary shapes. ACM Transactions on Graphics (TOG) 43 (2024), 1 -- 14

    2024

  26. [34]

    Xavier P. G. : Fast swept-volume distance for robust collision detection. Proceedings of International Conference on Robotics and Automation 2 (1997), 1162--1169 vol.2

    1997

  27. [35]

    : Surfacevoronoi: Efficiently computing voronoi diagrams over mesh surfaces with arbitrary distance solvers

    Xin S., Wang P., Xu R., Yan D., Chen S., Wang W., Zhang C., Tu C. : Surfacevoronoi: Efficiently computing voronoi diagrams over mesh surfaces with arbitrary distance solvers. ACM Trans. Graph. 41, 6 (Nov. 2022). https://doi.org/10.1145/3550454.3555453 doi:10.1145/3550454.3555453

    2022

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.