arxiv: 2605.14960 · v1 · submitted 2026-05-14 · 💻 cs.GR · cs.CG· cs.CV

Recognition: 1 theorem link

· Lean Theorem

Meschers: Geometry Processing of Impossible Objects

Ana Dodik , Isabella Yu , Kartik Chandra , Jonathan Ragan-Kelley , Joshua Tenenbaum , Vincent Sitzmann , Justin Solomon

Authors on Pith no claims yet

Pith reviewed 2026-05-15 03:05 UTC · model grok-4.3

classification 💻 cs.GR cs.CGcs.CV

keywords impossible objectsmeshesdiscrete exterior calculusgeometry processinginverse renderingEscher figurescomputer graphics

0 comments

The pith

Meschers represent impossible objects as meshes that support standard geometry processing and inverse rendering.

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

The paper introduces Meschers as a mesh representation for impossible objects like those in Escher drawings. Previous methods embed these objects in 3D by cutting them or bending along depth, which alters local geometry, disrupts smoothing, and can invalidate operations such as distance computation or relighting. Meschers instead build on discrete exterior calculus to keep the representation consistent for these tasks. This allows direct application of geometry processing algorithms and opens the door to inverse-rendering impossible objects. A reader would care because it provides a practical computational handle on objects that humans perceive but that cannot physically exist.

Core claim

Meschers are meshes capable of representing impossible constructions akin to those found in M.C. Escher's woodcuts. Our representation has a theoretical foundation in discrete exterior calculus and supports the use-cases above, as we demonstrate in a number of example applications. Moreover, because we can do discrete geometry processing on our representation, we can inverse-render impossible objects. We also compare our representation to cut and bend representations of impossible objects.

What carries the argument

Meschers, meshes grounded in discrete exterior calculus that encode impossible objects while preserving validity of standard operations such as distance computation and smoothing.

If this is right

Standard geometry processing operations such as smoothing and distance computation remain valid on the represented impossible objects.
Inverse rendering of impossible objects becomes possible using the same mesh representation.
Local geometry stays unchanged at any point, unlike cut-based embeddings that alter structure at seams.
Relighting and other graphics pipelines can be applied directly without the depth-bending artifacts of prior methods.

Where Pith is reading between the lines

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

The approach could support animation of impossible objects by applying time-varying geometry operators while maintaining perceptual consistency.
It may allow computational experiments that test which impossible figures remain stable under repeated smoothing or subdivision.
Integration with existing mesh libraries would let artists import Escher-style figures and run off-the-shelf tools without manual fixes.

Load-bearing premise

Discrete exterior calculus can be extended to meshes of impossible objects while keeping operations like smoothing and distance computation consistent without new artifacts.

What would settle it

Running a standard smoothing or distance algorithm on a Mescher and obtaining local geometric inconsistencies or lighting artifacts that do not match the intended impossible figure.

Figures

Figures reproduced from arXiv: 2605.14960 by Ana Dodik, Isabella Yu, Jonathan Ragan-Kelley, Joshua Tenenbaum, Justin Solomon, Kartik Chandra, Vincent Sitzmann.

**Figure 1.** Figure 1: The Impossibagel. The mescher is a geometry representation that allows rendering and relighting impossible objects (left), as well as performing intrinsic geometry processing operations like heat diffusion (center) and geodesic distance queries (right). Impossible objects, geometric constructions that humans can perceive but that cannot exist in real life, have been a topic of intrigue in visual arts, perc… view at source ↗

**Figure 2.** Figure 2: A mescher is defined by a triangle mesh where vertices are stored as [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: Beautiful View. Using the heat method, we can compute geodesic distances on meschers (Section 4.3). Here, we show the distance measured from the front steps. with a vibrant community of artists and a plethora of video and textual guides [Elber 2011; Gárate 2020, 2023], books [Ernst 1986], as well as online libraries containing hundreds of different impossible objects [Alexeev 2001; Elber 2012]. Other appro… view at source ↗

**Figure 4.** Figure 4: Window. This mescher is globally integrable: naïvely rendering it makes the bars intersect at the same depth (left). To create an impossibility, an additional form of depth ordering is necessary (Section 3.4). After introducing depth ordering, we can render Window as an impossible object. Here, we further subdivide the mescher and apply our smoothing operator to create specular highlights that emphasize t… view at source ↗

**Figure 5.** Figure 5: Impawssible Dog. Meschers support relighting. Here, we show the same mescher rendered with four different lighting conditions. This shows that some lighting conditions create a stronger illusory percept than others. camera assumption. Since the 4 new triangles are similar to the original triangle with a scaling factor of one half, a subdivided edge’s 𝜻 will equal to the one half of the 𝜻 of the edge to wh… view at source ↗

**Figure 7.** Figure 7: demonstrates a proof-of-concept of our inverse rendering pipeline. We initialize our shape to a standard (i.e., possible) torus which is then optimized to look more like the impossible triangle. Finally, we can check that we are indeed recovering an impossible object by examining the mescher’s harmonic component [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗

**Figure 8.** Figure 8: Mission: Impossible, Mission: Accomplished. Here, we compare our method to the two classes of existing methods for rendering impossible geometry. While all three methods support rendering, only our method supports a full set of rendering and geometry processing operations. The bent representation leads to artifacts when relighting, due to the perturbed normals. Both representations lead to artifacts when s… view at source ↗

**Figure 9.** Figure 9: Impossible Geometric Data Processing. Just like in ordinary meshes, meschers can be rendered out with a variety of shading techniques, including both flat (top) and smooth (bottom) shading. Ana Dodik acknowledges the generous support of the MIT Presidential Fellowship and the Mathworks Fellowship. Kartik Chandra acknowledges the Hertz Foundation and the NSF GRFP for funding. Jonathan Ragan-Kelley acknowle… view at source ↗

read the original abstract

Impossible objects, geometric constructions that humans can perceive but that cannot exist in real life, have been a topic of intrigue in visual arts, perception, and graphics, yet no satisfying computer representation of such objects exists. Previous work embeds impossible objects in 3D, cutting them or twisting/bending them in the depth axis. Cutting an impossible object changes its local geometry at the cut, which can hamper downstream graphics applications, such as smoothing, while bending makes it difficult to relight the object. Both of these can invalidate geometry operations, such as distance computation. As an alternative, we introduce Meschers, meshes capable of representing impossible constructions akin to those found in M.C. Escher's woodcuts. Our representation has a theoretical foundation in discrete exterior calculus and supports the use-cases above, as we demonstrate in a number of example applications. Moreover, because we can do discrete geometry processing on our representation, we can inverse-render impossible objects. We also compare our representation to cut and bend representations of impossible objects.

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

Meschers give a DEC-based mesh for impossible objects that supports standard geometry ops and inverse rendering without the local changes from cutting or bending, but the abstract leaves the operator details and consistency checks thin.

read the letter

The main thing here is that the authors built Meschers as meshes for impossible objects using discrete exterior calculus. This lets them run normal geometry processing like smoothing and distance computation, plus inverse rendering, while avoiding the geometry breaks that come from cutting the object or the relighting problems from bending it in depth. They compare it directly to those baselines and show example applications in graphics pipelines. That positioning and the demo work are the clearest strengths. The paper does a straightforward job explaining why prior embeddings fall short for downstream tasks and why a new representation matters for perception and rendering work. The applications feel practical rather than just illustrative. The soft spots sit in the theoretical claims. The abstract states the foundation in discrete exterior calculus and that standard operations remain valid, but it does not show the actual operator definitions, how inconsistencies at impossible junctions are handled, or quantitative checks on error after smoothing or distance queries. Without those steps visible, it is hard to judge whether the extension really preserves the calculus without new artifacts. The weakest assumption is that the discrete operators transfer cleanly; that needs explicit verification in the full text. This paper is for graphics researchers who already work with discrete geometry or inverse rendering and want to handle non-physical objects inside existing pipelines. A reader focused on perceptual rendering or Escher-style geometry would get the most from the examples and comparisons. It deserves a serious referee to check the derivations and run the numbers on the operations, even if revisions are likely. I would send it out for review rather than desk reject.

Referee Report

2 major / 2 minor

Summary. The paper introduces Meschers, a mesh representation for impossible objects (e.g., Escher-like constructions) that cannot exist in Euclidean 3D space. It claims a foundation in discrete exterior calculus that preserves standard geometry-processing operations such as smoothing and distance computation, enables inverse rendering, and avoids the local-geometry changes or relighting difficulties of prior cut and bend embeddings. The work demonstrates several applications and includes comparisons to those earlier representations.

Significance. If the discrete-exterior-calculus extension is valid, the contribution would be significant for geometry processing and graphics: it would allow consistent operations on non-realizable geometries, support inverse problems, and provide a cleaner alternative to ad-hoc 3D embeddings. The absence of free parameters and the grounding in established DEC operators are strengths that could make the representation reusable across downstream tasks.

major comments (2)

[§3] §3 (Theoretical Foundation): the claim that standard DEC operators remain valid on Meschers requires explicit construction of the modified exterior derivative and Hodge star; without the precise operator definitions and a proof that they commute with the impossibility constraints, the preservation of distance computation and smoothing cannot be verified.
[§5] §5 (Results and Comparisons): the quantitative comparison to cut and bend methods reports only qualitative visual differences; adding error metrics (e.g., Hausdorff distance after smoothing or gradient consistency before/after inverse rendering) is needed to substantiate the claim that Meschers “invalidates geometry operations” less than the baselines.

minor comments (2)

[Figure 4] Figure 4 caption: the legend for the inverse-rendered examples is missing; readers cannot distinguish the Mescher result from the cut/bend baselines without it.
[Notation] Notation: the symbol for the Mescher-specific 2-form is introduced without a clear mapping to the standard DEC notation used in the rest of the paper; a short table of correspondences would improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive assessment and recommendation for minor revision. The comments highlight useful ways to strengthen the presentation of the theoretical foundation and the empirical comparisons. We address each major comment below.

read point-by-point responses

Referee: [§3] §3 (Theoretical Foundation): the claim that standard DEC operators remain valid on Meschers requires explicit construction of the modified exterior derivative and Hodge star; without the precise operator definitions and a proof that they commute with the impossibility constraints, the preservation of distance computation and smoothing cannot be verified.

Authors: We agree that explicit operator definitions will improve clarity and verifiability. In the revised manuscript we will add the precise constructions of the modified exterior derivative and Hodge star for Meschers, together with a short proof that these operators commute with the impossibility constraints (i.e., that the discrete Stokes theorem and the usual DEC identities continue to hold on the augmented mesh). This addition will directly support the claims about distance computation and smoothing. revision: yes
Referee: [§5] §5 (Results and Comparisons): the quantitative comparison to cut and bend methods reports only qualitative visual differences; adding error metrics (e.g., Hausdorff distance after smoothing or gradient consistency before/after inverse rendering) is needed to substantiate the claim that Meschers “invalidates geometry operations” less than the baselines.

Authors: We accept that quantitative metrics would strengthen the comparison. In the revision we will report Hausdorff distances between the original and smoothed meshes for all three representations, as well as a gradient-consistency measure (L2 norm of the difference between rendered and target gradients) before and after inverse rendering. These numbers will be added to the existing visual comparisons in §5. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper presents Meschers as a representation grounded in established discrete exterior calculus, with the abstract asserting a theoretical foundation that supports standard geometry operations like distance computation and smoothing. No derivations, equations, or self-citations are visible in the provided text that reduce any central claim to a fitted parameter, self-definition, or load-bearing prior result by the same authors. The extension to impossible objects is framed as preserving validity of existing operators rather than redefining them circularly, and no uniqueness theorems or ansatzes are imported in a way that collapses the argument to its own inputs. This is a self-contained claim against external benchmarks with no enumerated circularity patterns exhibited.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the applicability of discrete exterior calculus to inconsistent geometric objects; no free parameters or invented physical entities are mentioned in the abstract.

axioms (1)

domain assumption Discrete exterior calculus can be adapted to represent impossible objects while preserving validity of geometry operations
Invoked in the abstract as the theoretical foundation for the Mescher representation.

invented entities (1)

Meschers no independent evidence
purpose: Mesh representation for impossible objects
New mesh type introduced to encode impossible constructions without cutting or bending.

pith-pipeline@v0.9.0 · 5490 in / 1198 out tokens · 32818 ms · 2026-05-15T03:05:58.675967+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/AlexanderDuality.lean alexander_duality_circle_linking unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

the local integrability condition can be expressed as a linear constraint: d12 ζ=0. ... ζ=d01 z + ω (harmonic component)

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

47 extracted references · 47 canonical work pages

[1]

In ACM SIGGRAPH 2006 Papers

Photographing long scenes with multi-viewpoint panoramas. In ACM SIGGRAPH 2006 Papers. 853–861. Maneesh Agrawala, Denis Zorin, and Tamara Munzner

work page 2006
[2]

InRendering Techniques 2000: Proceedings of the Eurographics Workshop in Brno, Czech Republic, June 26–28, 2000

Artistic multiprojection rendering. InRendering Techniques 2000: Proceedings of the Eurographics Workshop in Brno, Czech Republic, June 26–28, 2000

work page 2000
[3]

The Visual Computer19, 2 (2003), 105–114

Differential coordinates for local mesh morphing and deformation. The Visual Computer19, 2 (2003), 105–114. Vlad Alexeev. 2001.Impossible world — im-possible.info. [Accessed 16-04-2025]. Irving Biederman

work page 2003
[4]

Computer vision, graphics, and image processing32, 1 (1985), 29–73

Human image understanding: Recent research and a theory. Computer vision, graphics, and image processing32, 1 (1985), 29–73. David Bommes, Henrik Zimmer, and Leif Kobbelt. 2023.Mixed-Integer Quadrangulation (1 ed.). Association for Computing Machinery, New York, NY, USA. https://doi. org/10.1145/3596711.3596740 Ryan Burgert, Xiang Li, Abe Leite, Kanchana ...

work page doi:10.1145/3596711.3596740 1985
[5]

Ryan Capouellez, Jiacheng Dai, Aaron Hertzmann, and Denis Zorin

Diffusion illusions: Hiding images in plain sight.arXiv preprint arXiv:2312.03817 (2023). Ryan Capouellez, Jiacheng Dai, Aaron Hertzmann, and Denis Zorin

work page arXiv 2023
[6]

InACM SIGGRAPH 2023 Conference Proceedings(Los Angeles, CA, USA)(Siggraph ’23)

Algebraic Smooth Occluding Contours. InACM SIGGRAPH 2023 Conference Proceedings(Los Angeles, CA, USA)(Siggraph ’23). Association for Computing Machinery, New York, NY, USA, Article 39, 10 pages. https://doi.org/10.1145/3588432.3591547 Patrick Cavanagh

work page doi:10.1145/3588432.3591547 2023
[7]

Kartik Chandra, Tzu-Mao Li, Joshua Tenenbaum, and Jonathan Ragan-Kelley

The artist as neuroscientist.Nature434, 7031 (2005), 301–307. Kartik Chandra, Tzu-Mao Li, Joshua Tenenbaum, and Jonathan Ragan-Kelley

work page 2005
[8]

InACM SIGGRAPH 2022 Conference Proceedings

Designing perceptual puzzles by differentiating probabilistic programs. InACM SIGGRAPH 2022 Conference Proceedings. 1–9. Ming-Te Chi, Tong-Yee Lee, Yingge Qu, and Tien-Tsin Wong

work page 2022
[9]

Self-Animating Images: Illusory Motion Using Repeated Asymmetric Patterns.ACM Trans. Graph. 27, 3 (aug 2008), 1–8. https://doi.org/10.1145/1360612.1360661 Hung-Kuo Chu, Wei-Hsin Hsu, Niloy J Mitra, Daniel Cohen-Or, Tien-Tsin Wong, and Tong-Yee Lee

work page doi:10.1145/1360612.1360661 2008
[10]

Graph.29, 4 (2010), 51–1

Camouflage images.ACM Trans. Graph.29, 4 (2010), 51–1. https://dl.acm.org/doi/abs/10.1145/1833349.1778788 Omar Cornut

work page doi:10.1145/1833349.1778788 2010
[11]

InACM SIGGRAPH 2013 Courses (Anaheim, California)(Siggraph ’13)

Digital geometry processing with discrete exterior calculus. InACM SIGGRAPH 2013 Courses (Anaheim, California)(Siggraph ’13). Association for Computing Machinery, New York, NY, USA, Article 7, 126 pages. https://doi.org/10.1145/2504435.2504442 Keenan Crane, Clarisse Weischedel, and Max Wardetzky

work page doi:10.1145/2504435.2504442 2013
[12]

ACM60, 11 (Oct

The Heat Method for Distance Computation.Commun. ACM60, 11 (Oct. 2017), 90–99. https://doi.org/ 10.1145/3131280 Fernando de Goes, Mathieu Desbrun, Mark Meyer, and Tony DeRose

work page doi:10.1145/3131280 2017
[13]

Graph.35, 4, Article 133 (July 2016), 11 pages

Subdivision exterior calculus for geometry processing.ACM Trans. Graph.35, 4, Article 133 (July 2016), 11 pages. https://doi.org/10.1145/2897824.2925880 Mathieu Desbrun, Eva Kanso, and Yiying Tong

work page doi:10.1145/2897824.2925880 2016
[14]

InACM SIGGRAPH 2006 Courses(Boston, Massachusetts) (Siggraph ’06)

Discrete differential forms for computational modeling. InACM SIGGRAPH 2006 Courses(Boston, Massachusetts) (Siggraph ’06). Association for Computing Machinery, New York, NY, USA, 39–54. https://doi.org/10.1145/1185657.1185665 Mathieu Desbrun, Melvin Leok, and Jerrold E. Marsden

work page doi:10.1145/1185657.1185665 2006
[15]

Applied Numerical Mathematics53, 2-4 (May 2005), 231–248

Discrete Poincaré Lemma. Applied Numerical Mathematics53, 2-4 (May 2005), 231–248. https://doi.org/10. 1016/j.apnum.2004.09.035 Mathieu Desbrun, Mark Meyer, Peter Schroder, and Alan H. Barr. 2023.Implicit Fairing of Irregular Meshes using Diffusion and Curvature Flow(1 ed.). Association for Com- puting Machinery, New York, NY, USA. https://doi.org/10.1145...

work page doi:10.1145/3596711.3596729 2005
[16]

https://doi.org/10.1068/p070283 arXiv:https://doi.org/10.1068/p070283 Pmid: 693228

The Penrose Triangle and a Family of Related Fig- ures.Perception7, 3 (1978), 283–296. https://doi.org/10.1068/p070283 arXiv:https://doi.org/10.1068/p070283 Pmid: 693228. Gershon Elber

work page doi:10.1068/p070283 1978
[17]

Gershon Elber

Modeling (seemingly) impossible models.Computers & Graphics 35, 3 (2011), 632–638. Gershon Elber. 2012.Escher for Real — gershonelber.org. [Accessed 16-04-2025]. B. Ernst. 1986.Adventures with Impossible Figures. Tarquin. https://books.google.com/ books?id=HU_vAAAAMAAJ M.C. Escher. 1961.The Graphic Work of M.C. Escher. Oldbourne. https://books.google. com...

work page 2011
[18]

Erez Freud, Tzvi Ganel, and Galia Avidan

Motion without movement.ACM Siggraph Computer Graphics25, 4 (1991), 27–30. Erez Freud, Tzvi Ganel, and Galia Avidan

work page 1991
[19]

https://doi.org/10.1016/j.neuroimage.2012.08.070 Erez Freud, Bat-Sheva Hadad, Galia Avidan, and Tzvi Ganel

Representation of possible and impossi- ble objects in the human visual cortex: Evidence from fMRI adaptation.NeuroImage 64 (2013), 685–692. https://doi.org/10.1016/j.neuroimage.2012.08.070 Erez Freud, Bat-Sheva Hadad, Galia Avidan, and Tzvi Ganel

work page doi:10.1016/j.neuroimage.2012.08.070 2013
[20]

https://doi.org/10.3389/fpsyg.2015.00094 Matías Gárate

Evidence for similar early but not late representation of possible and impossible objects.Frontiers in Psychology6 (2015). https://doi.org/10.3389/fpsyg.2015.00094 Matías Gárate. 2020.Art Spotlight: Impossible Penrose Snake — sketchfab.com. [Accessed 16-04-2025]. Matías Gárate. 2023.The ins and outs of modelling impossible figures. — Blender Confer- ence ...

work page doi:10.3389/fpsyg.2015.00094 2015
[21]

Daniel Geng, Inbum Park, and Andrew Owens

Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Models.arXiv preprint arXiv:2311.17919 (2023). Daniel Geng, Inbum Park, and Andrew Owens

work page arXiv 2023
[22]

Factorized Diffusion: Perceptual Illusions by Noise Decomposition.arXiv preprint arXiv:2404.11615(2024). R. L. (Richard Langton) Gregory. 1970.The intelligent eye, by R. L. Gregory.Weidenfeld & Nicolson, London. ACM Trans. Graph., Vol. 44, No. 4, Article . Publication date: August

work page arXiv 2024
[23]

https: //doi.org/10.1016/j.visres.2021.09.004 Aaron Hertzmann

A failure to learn object shape geometry: Implications for convolutional neural networks as plausible models of biological vision.Vision Research189 (2021), 81–92. https: //doi.org/10.1016/j.visres.2021.09.004 Aaron Hertzmann

work page doi:10.1016/j.visres.2021.09.004 2021
[24]

https://doi.org/10.1167/jov.24

Toward a theory of perspective perception in pic- tures.Journal of Vision24, 4 (04 2024), 23–23. https://doi.org/10.1167/jov.24. 4.23 arXiv:https://arvojournals.org/arvo/content_public/journal/jov/938670/i1534- 7362-24-4-23_1713961163.65834.pdf Anil N. Hirani. 2003.Discrete Exterior Calculus. Ph. D. Dissertation. California Institute of Technology. http:/...

work page doi:10.1167/jov.24 2024
[25]

InProceedings of Bridges 2014: Mathe- matics, Music, Art, Architecture, Culture, Gary Greenfield, George Hart, and Reza Sarhangi (Eds.)

Constructing Drawings of Impossible Figures with Axonomet- ric Blocks and Pseudo-3D Manipulations. InProceedings of Bridges 2014: Mathe- matics, Music, Art, Architecture, Culture, Gary Greenfield, George Hart, and Reza Sarhangi (Eds.). Tessellations Publishing, Phoenix, Arizona, 159–166. http://archive. bridgesmathart.org/2014/bridges2014-159.html Olga A ...

work page 2014
[26]

InACM SIGGRAPH 2006 Papers

Smoothsketch: 3d free-form shapes from complex sketches. InACM SIGGRAPH 2006 Papers. 589–598. Chih W Khoh and Peter Kovesi

work page 2006
[27]

Graph.34 (2015)

Stripe Patterns on Surfaces.ACM Trans. Graph.34 (2015). Issue

work page 2015
[28]

Series A: Mathematical, Physical and Engineering Sciences356, 1740 (1998), 1071–1086

Pictorial relief.Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences356, 1740 (1998), 1071–1086. Felix Kälberer, Matthias Nieser, and Konrad Polthier

work page 1998
[29]

QuadCover - Sur- face Parameterization using Branched Coverings.Computer Graphics Fo- rum26, 3 (2007), 375–384. https://doi.org/10.1111/j.1467-8659.2007.01060.x arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1467-8659.2007.01060.x Chi-Fu William Lai, Sai-Kit Yeung, Xiaoqi Yan, Chi-Wing Fu, and Chi-Keung Tang

work page doi:10.1111/j.1467-8659.2007.01060.x 2007
[30]

IEEE transactions on visualization and computer graphics22, 10 (2015), 2275–2288

3D navigation on impossible figures via dynamically reconfigurable maze. IEEE transactions on visualization and computer graphics22, 10 (2015), 2275–2288. Tzu-Mao Li, Miika Aittala, Frédo Durand, and Jaakko Lehtinen

work page 2015
[31]

Yuanbo Li, Tianyi Ma, Zaineb Aljumayaat, and Daniel Ritchie

Differentiable monte carlo ray tracing through edge sampling.ACM Transactions on Graphics (TOG)37, 6 (2018), 1–11. Yuanbo Li, Tianyi Ma, Zaineb Aljumayaat, and Daniel Ritchie

work page 2018
[32]

In Proceedings Shape Modeling Applications, 2004.Ieee, 181–190

Differential coordinates for interactive mesh editing. In Proceedings Shape Modeling Applications, 2004.Ieee, 181–190. Shichen Liu, Tianye Li, Weikai Chen, and Hao Li

work page 2004
[33]

Margaret S Livingstone

Soft Rasterizer: A Differentiable Renderer for Image-based 3D Reasoning.The IEEE International Conference on Computer Vision (ICCV)(Oct 2019). Margaret S Livingstone. 2022.Vision and art (updated and expanded edition). Abrams. Charles T. Loop

work page 2019
[34]

https://cg.cs.tsinghua.edu.cn/papers/TVCG-2013-changeblindness.pdf David Marr

Change blind- ness images.IEEE transactions on visualization and computer graphics19, 11 (2013), 1808–1819. https://cg.cs.tsinghua.edu.cn/papers/TVCG-2013-changeblindness.pdf David Marr. 1982.Vision: A computational investigation into the human representation and processing of visual information. MIT press. Baptiste Nicolet, Alec Jacobson, and Wenzel Jakob

work page 2013
[35]

Yu Okano, Shogo Fukushima, Masahiro Furukawa, and Hiroyuki Kajimoto

Large steps in inverse rendering of geometry.ACM Transactions on Graphics (TOG)40, 6 (2021), 1–13. Yu Okano, Shogo Fukushima, Masahiro Furukawa, and Hiroyuki Kajimoto

work page 2021
[36]

InACM SIGGRAPH ASIA 2010 Posters(Seoul, Republic of Korea)(Sa ’10)

Embedded Motion: Generating the Perception of Motion in Peripheral Vision. InACM SIGGRAPH ASIA 2010 Posters(Seoul, Republic of Korea)(Sa ’10). As- sociation for Computing Machinery, New York, NY, USA, Article 41, 1 pages. https://doi.org/10.1145/1900354.1900400 Aude Oliva, Antonio Torralba, and Philippe G Schyns

work page doi:10.1145/1900354.1900400 2010
[37]

https://stanford.edu/class/ ee367/reading/OlivaTorralb_Hybrid_Siggraph06.pdf Shigeru Owada and Jun Fujiki

Hybrid images.ACM Transactions on Graphics (TOG)25, 3 (2006), 527–532. https://stanford.edu/class/ ee367/reading/OlivaTorralb_Hybrid_Siggraph06.pdf Shigeru Owada and Jun Fujiki

work page 2006
[38]

Roger Penrose

Impossible objects: a special type of visual illusion.British Journal of Psychology(1958). Roger Penrose

work page 1958
[39]

https://api.semanticscholar.org/CorpusID:125905129 Nikhila Ravi, Jeremy Reizenstein, David Novotny, Taylor Gordon, Wan-Yen Lo, Justin Johnson, and Georgia Gkioxari

On the Cohomology of Impossible Figures.Leonardo25 (1993), 245–247. https://api.semanticscholar.org/CorpusID:125905129 Nikhila Ravi, Jeremy Reizenstein, David Novotny, Taylor Gordon, Wan-Yen Lo, Justin Johnson, and Georgia Gkioxari

work page 1993
[40]

Accelerating 3d deep learning with pytorch3d.arXiv preprint arXiv:2007.08501, 2020

Accelerating 3D Deep Learning with Py- Torch3D.arXiv:2007.08501(2020). Oscar Reutersvärd

work page arXiv 2007
[41]

Guillermo Savransky, Dan Dimerman, and Craig Gotsman

How to make impossible objects possible: Anamorphic deformation of textured NURBS.Computer Aided Geometric Design78 (2020), 101826. Guillermo Savransky, Dan Dimerman, and Craig Gotsman

work page 2020
[42]

Herbert A

A new ambiguous figure: A threestick clevis.The American Journal of Psychology(1964). Herbert A. Simon

work page 1964
[43]

https://doi.org/10.1111/j.2044-8295.1967.tb01051.x arXiv:https://bpspsychub.onlinelibrary.wiley.com/doi/pdf/10.1111/j.2044- 8295.1967.tb01051.x Olga Sorkine

An Information-processing Explanation Of Some Perceptual Phenomena.British Journal of Psychology58, 1- 2 (1967), 1–12. https://doi.org/10.1111/j.2044-8295.1967.tb01051.x arXiv:https://bpspsychub.onlinelibrary.wiley.com/doi/pdf/10.1111/j.2044- 8295.1967.tb01051.x Olga Sorkine

work page doi:10.1111/j.2044-8295.1967.tb01051.x 1967
[44]

Benjamin Adam Taylor

Three-dimensional realization of anomalous pictures—An application of picture interpretation theory to toy design.Pattern Recognition30, 7 (1997), 1061–1067. Benjamin Adam Taylor. 2020.Modeling and Rendering Three-Dimensional Impossible Objects. Bangor University (United Kingdom). E. Térouanne

work page 1997
[45]

https://doi.org/10.1016/S0166- 4115(08)62082-8 Xi Wang, Zoya Bylinskii, Aaron Hertzmann, and Robert Pepperell

North-Holland, 105–120. https://doi.org/10.1016/S0166- 4115(08)62082-8 Xi Wang, Zoya Bylinskii, Aaron Hertzmann, and Robert Pepperell

work page doi:10.1016/s0166-
[46]

Ethan Weber*, Riley Peterlinz*, Rohan Mathur, Frederik Warburg, Alexei A

Toward quantifying ambiguities in artistic images.ACM Transactions on Applied Perception (TAP)17, 4 (2020), 1–10. Ethan Weber*, Riley Peterlinz*, Rohan Mathur, Frederik Warburg, Alexei A. Efros, and Angjoo Kanazawa

work page 2020
[47]

ACM Trans

Model- ing and rendering of impossible figures.ACM Transactions on Graphics (ToG)29, 2 (2010), 1–15. ACM Trans. Graph., Vol. 44, No. 4, Article . Publication date: August 2025

work page 2010