Pith. sign in

REVIEW 3 major objections 2 minor 1 cited by

Polarization-Resolved Chlorophyll Imaging for Non-Invasive Plant Tissue Assessment Using a Silicon-Rich Nitride Metalens Array

T0 review · 3 major / 2 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read This paper claims a compact silicon-rich nitride metalens array can image plant tissue in two orthogonal polarizations at the 660 nm chlorophyll absorption peak.

desk verdict The abstract describes a metalens plant-imaging device, but the attached full text is an unrelated CLIP symmetry-detection paper; there is nothing to referee. read the letter →

arxiv 2508.14191 v1 pith:6F4FTTKF submitted 2025-08-19 physics.optics physics.bio-ph

Alireza Khalilian , Mehdi Sh. Yeganeh , Konstantin Tsoi , Bowen Yu , Joe Lo , Yasha Yi This is my paper · ORCID
classification physics.opticsphysics.bio-ph
keywords metalensarraypolarization-resolvedimagingchlorophyllabsorptionplantstressdetectionsilicon-richnitrideCMOS-compatiblenanophotonicslabel-freebirefringentmeta-atoms
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 seeks to show that bulky polarization optics can be replaced by a flat, mass-producible metalens array for plant imaging. It claims simultaneous X/Y polarization transmission images at the 660 nm chlorophyll absorption peak, with polarization fusion and difference maps that reveal tissue anisotropy and pigment variation in healthy and stressed leaves. If the device works as described, plant stress assessment could become real-time, label-free, and compact. The supplied full text, however, is an unrelated manuscript on symmetry detection using a vision-language model; no fabrication, measurement, or validation data for the metalens appear in this submission.

What carries the argument

The central object is a silicon-rich nitride (SRN) metalens array: a flat, CMOS-compatible optical element whose birefringent meta-atoms—nanoscale pillars that impart different phase delays to X- versus Y-polarized light—are arranged by inverse design to focus the two polarizations onto orthogonally sensitive imaging channels. The array works at 660 nm, matching chlorophyll absorption, and its role is to replace bulky polarizers, waveplates, and rotating optics with a single flat component that captures both polarization channels at once. Polarization fusion and difference maps are then the output signals that encode structural anisotropy and pigment variation in leaves.

What would settle it

Expose leaves to controlled stress and measure polarization difference maps with both the metalens array and a conventional polarizing microscope, comparing against an independent stress assay; if the maps do not track the independent stress markers, or if they vanish when the sample is rotated relative to the lens axes, the central claim is refuted.

Watch

Extended reading notes

Core claim

The paper claims that a silicon-rich nitride metalens array with orthogonally sensitive elements can simultaneously capture X- and Y-linearly polarized transmission images of plant tissue at 660 nm, and that fusing or differencing those channels reveals structural anisotropy and pigment variation that distinguishes healthy from stressed leaves. The claimed mechanism is an inverse-designed, birefringent meta-atom lens with high numerical aperture, high transmission, and spectral alignment with chlorophyll absorption, fabricated in CMOS-compatible processes. The stated value is a compact, integrated substitute for conventional polarization-resolved imaging systems, with potential for agricultu

Load-bearing premise

The device must actually have been built and measured as the abstract implies, and the polarization difference must reflect the leaf's physiological state rather than its geometry, preparation, or the lens's own polarization artifacts.

Editorial extensions

If this is right

  • Polarization-resolved plant imaging could be performed with a compact flat optic rather than a microscope fitted with rotating polarizers.
  • Simultaneous capture of two orthogonal polarization channels removes moving parts, enabling real-time observation of leaf tissue.
  • Operation at 660 nm ties the image contrast directly to chlorophyll absorption, so stress responses that alter pigment content or chloroplast organization could be visible without staining.
  • CMOS-compatible fabrication means the array could potentially be manufactured at scale in existing semiconductor facilities.
  • If the polarization difference maps are reproducible, the device could serve as a low-cost field sensor for agricultural stress screening.

Reading between the lines

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

  • Editorial inference: the body text of this submission is an unrelated paper on symmetry detection with a vision-language model, so the metalens abstract currently stands without supporting methods or data; that missing evidence is the main obstacle to taking the claim as established.
  • A testable extension: image a leaf before and after inducing water or heat stress; if the polarization difference maps track independent stress markers such as chlorophyll fluorescence or leaf water potential, the physiological interpretation is supported.
  • Another testable extension: rotate the sample relative to the lens axes and check whether the X/Y difference rotates with the leaf; if it follows the sample rather than the fixed lens axes, it likely reflects tissue anisotropy rather than lens artifacts.
  • If the polarization contrast truly sits at 660 nm, the array could be combined with other narrowband channels to separate pigment absorption effects from birefringent structural effects in the same leaf.
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

3 major / 2 minor

Summary. The arXiv submission, as identified by its abstract, claims a silicon-rich nitride (SRN) metalens array for polarization-resolved, label-free imaging of plant tissue at 660 nm, with simultaneous X/Y linear polarization capture, inverse-designed birefringent meta-atoms, CMOS-compatible fabrication, and demonstration of stress-induced structural anisotropy in leaves. However, the supplied full text is the paper "CLIPSym: Delving into Symmetry Detection with CLIP" (Yang, Rahman, Yeh, Purdue), a computer vision paper on symmetry detection using vision-language models. The full text contains no occurrence of metalens, chlorophyll, polarization-resolved imaging, 660 nm, birefringent meta-atoms, leaf stress, optical characterization, or any experimental methods or results relevant to the abstract's claims. Thus the manuscript body provides zero support for the claimed device and its application.

Significance. If the claims in the abstract were substantiated, the work could be significant for compact, mass-producible polarization imaging in plant biology and agriculture, potentially enabling low-cost stress phenotyping. However, the submitted manuscript contains none of the required evidence: no design methodology, simulation results, fabrication details, transmission spectra, numerical aperture measurements, polarization extinction ratios, imaging data, or stress-assay controls. The only support for the central claims is the abstract itself. Consequently, the scientific contribution cannot be evaluated, and the significance is entirely conditional on evidence that is not present in this submission.

major comments (3)
  1. [Full Text (CLIPSym)] The full text is an unrelated paper on symmetry detection with CLIP. It contains no mention of metalenses, silicon-rich nitride, birefringent meta-atoms, 660 nm operation, chlorophyll imaging, or plant stress. None of the abstract's experimental claims—large numerical aperture, high transmission, simultaneous X/Y polarization imaging, or real-time stress assessment—are supported by any methods, data, or analysis in the body. This is a fundamental verifiability failure: the manuscript as submitted does not contain the claimed research.
  2. [Abstract] The abstract states that the SRN metalens 'achieves a large numerical aperture, high transmission, and spectral alignment with biological absorbers.' No quantitative values, error bars, comparison standards, or measurement protocols are provided anywhere in the manuscript. Without device characterization (NA, transmission spectrum, polarization crosstalk, fabrication yield) the central optical performance claims are uncheckable.
  3. [Abstract] The claim that 'Polarization fusion and difference mapping reveal structural anisotropy and pigment variation in both healthy and stressed leaves' requires a controlled stress assay and a demonstration that the polarization signal tracks stress rather than leaf thickness, venation, preparation artifacts, or metalens polarization crosstalk. No images, ground-truth labels, statistical comparisons, or control experiments are presented. The physiological interpretation is therefore unsupported.
minor comments (2)
  1. [Submission metadata] The arXiv category (physics.optics) and the abstract do not match the full text, which is a computer vision paper. The submission appears to have a substantial mismatch between the abstract and the body.
  2. [Full Text] The full text contains no figures, tables, or equations relevant to the abstract. The only content is the CLIPSym paper, its references, and an appendix on symmetry detection. This makes even partial assessment of the claimed optics impossible.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found; the manuscript body is an unrelated paper, so the abstract's claims are unsupported rather than circular.

full rationale

The circularity pass looks for load-bearing derivations that reduce to their own inputs, fitted parameters renamed as predictions, or self-citations carrying the argument. Here there is no derivation chain to audit: the abstract of arXiv:2508.14191 claims a silicon-rich nitride metalens array with inverse-designed birefringent meta-atoms, 660 nm polarization-resolved leaf imaging, and stress assessment, but the supplied full text is an unrelated computer-vision paper, 'CLIPSym: Delving into Symmetry Detection with CLIP,' by different authors on CLIP-based symmetry detection. The full text contains no metalens, no chlorophyll, no polarization-resolved imaging, no optical simulation, no fabrication, no transmission spectra, no NA measurements, and no leaf-stress experiment. Because the claimed derivation and experimental results are entirely absent, the central assertions are unsupported and unverifiable; however, absence of support is not circularity. There is no equation that reduces to its own input, no fitted parameter later relabeled as a prediction, and no self-citation chain forcing the conclusion. The appropriate finding is therefore score 0 for circularity, with the caveat that the manuscript suffers from a fundamental verifiability failure: the body does not address the abstract in any way. This is not a case where a self-contained derivation is present, so the default low-circularity score applies, but the paper's scientific claims cannot be checked at all.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The central claim rests on several unverified domain assumptions because the supplied full text does not contain the corresponding experiments. These are listed below; no free parameters or invented entities are identifiable from the abstract.

assumptions (3)
  • domain assumption Polarization contrast in leaves correlates with physiological stress.
    The abstract reads stress responses off polarization difference maps, but no stress assay, calibration, or biological ground truth is provided.
  • domain assumption The SRN meta-atoms at 660 nm preserve linear polarization with sufficient extinction ratio.
    The two orthogonal lens channels must not bleed into each other; the abstract reports transmission but not extinction ratio or crosstalk.
  • domain assumption CMOS-compatible fabrication reproduces the inverse-designed meta-atom geometry.
    The feasibility claim depends on fabrication fidelity; no process parameters, SEM images, or spectral measurements are given.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Polarization-Resolved Chlorophyll Imaging for Non-Invasive Plant Tissue Assessment Using a Silicon-Rich Nitride Metalens Array." pith.science (2026). https://pith.science/paper/6F4FTTKF

@misc{pith2026250814191,
  author       = {Pith},
  title        = {Pith review of: Polarization-Resolved Chlorophyll Imaging for Non-Invasive Plant Tissue Assessment Using a Silicon-Rich Nitride Metalens Array},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6F4FTTKF}},
  note         = {Machine review of arXiv:2508.14191}
}
read the original abstract

Polarization-sensitive imaging enhances contrast and reveals structural features in biological tissues that are often missed by intensity-based methods, but its adoption is limited by bulky optics. We present a compact silicon-rich nitride (SRN) metalens array for high-resolution, polarization-resolved imaging of plant tissue at the chlorophyll absorption peak (660\,nm). The array integrates orthogonally sensitive metalenses to simultaneously capture X- and Y-linearly polarized transmission images, enabling real-time, label-free assessment of plant microstructure and stress responses. Polarization fusion and difference mapping reveal structural anisotropy and pigment variation in both healthy and stressed leaves. The SRN metalens, designed via an inverse approach using birefringent meta-atoms and fabricated through CMOS-compatible processes, achieves a large numerical aperture, high transmission, and spectral alignment with biological absorbers. This work demonstrates the feasibility of compact, integrated polarization-resolved imaging, offering a scalable alternative to conventional systems. The approach holds potential for biomedical and agricultural applications, where detecting subtle polarization-dependent changes could enable early diagnosis and tissue characterization

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. A Bayesian Semiparametric Mixture Model for Clustering Zero-Inflated Microbiome Data

    stat.ME 2025-08 unverdicted novelty 4.0 of 10

    A Bayesian semiparametric mixture model for zero-inflated compositional microbiome counts estimates the number of clusters and assigns subjects to them in one framework.

Reference graph

Works this paper leans on

78 extracted references · 70 canonical work pages · cited by 1 Pith paper

  1. [1]

    I. R. Atadjanov and S. Lee. Reflection symmetry detec- tion via appearance of structure descriptor. In ECCV,

    work page

  2. [2]

    Bahng, A

    H. Bahng, A. Jahanian, S. Sankaranarayanan, and P. Isola. Exploring visual prompts for adapting large- scale models. arXiv preprint arXiv:2203.17274, 2022. 2

    arXiv 2022

  3. [3]

    Cicconet, V

    M. Cicconet, V . Birodkar, M. Lund, M. Werman, and D. Geiger. A convolutional approach to reflection sym- metry. Pattern Recognition Letters, 2017. 1

    work page 2017

  4. [4]

    Cohen and M

    T. Cohen and M. Welling. Group equivariant convolu- tional networks. In ICML, 2016. 2, 4

    work page 2016

  5. [5]

    T. S. Cohen and M. Welling. Steerable CNNs. InICLR,

    work page

  6. [6]

    J. D. Delius and G. Habers. Symmetry: can pigeons conceptualize it? Behavioral biology, 1978. 1

    work page 1978

  7. [7]

    Dosovitskiy, L

    A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weis- senborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Min- derer, G. Heigold, S. Gelly, et al. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020. 3, 17

    arXiv 2010

  8. [8]

    Driver, G

    J. Driver, G. C. Baylis, and R. D. Rafal. Preserved figure-ground segregation and symmetry perception in visual neglect. Nature, 1992. 1

    work page 1992

Show all 78 references
  1. [9]

    Dumoulin, E

    V . Dumoulin, E. Perez, N. Schucher, F. Strub, H. d. Vries, A. Courville, and Y . Bengio. Feature-wise trans- formations. Distill, 2018. 3

    2018

  2. [10]

    Ecins, C

    A. Ecins, C. Fermüller, and Y . Aloimonos. Cluttered scene segmentation using the symmetry constraint. In ICRA, 2016. 1

    2016

  3. [11]

    Fukushima and M

    K. Fukushima and M. Kikuchi. Symmetry axis ex- traction by a neural network. Neurocomputing, 2006. 2

    2006

  4. [12]

    Funk and Y

    C. Funk and Y . Liu. Beyond planar symmetry: Mod- eling human perception of reflection and rotation sym- metries in the wild. In ICCV, 2017. 1, 2, 5, 6

    2017

  5. [13]

    C. Funk, S. Lee, M. R. Oswald, S. Tsogkas, W. Shen, A. Cohen, S. Dickinson, and Y . Liu. 2017 iccv chal- lenge: Detecting symmetry in the wild. In ICCVW,

    2017

  6. [14]

    Gnutti, F

    A. Gnutti, F. Guerrini, and R. Leonardi. Combining ap- pearance and gradient information for image symmetry detection. IEEE TIP, 2021. 2

    2021

  7. [15]

    Hartford, D

    J. Hartford, D. Graham, K. Leyton-Brown, and S. Ra- vanbakhsh. Deep models of interactions across sets. In Proc. ICML, 2018. 2

    2018

  8. [16]

    D. C. Hauagge and N. Snavely. Image matching using local symmetry features. In CVPR, 2012. 1

    2012

  9. [17]

    K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. In ICCV, 2016. 5

    2016

  10. [18]

    X. Hu, C. Zhang, Y . Zhang, B. Hai, K. Yu, and Z. He. Learning to adapt CLIP for few-shot monocular depth estimation. In WACV, 2024. 2

    2024

  11. [19]

    J. Je, J. Liu, G. Yang, B. Deng, S. Cai, G. Wetzstein, O. Litany, and L. Guibas. Robust symmetry detection via riemannian langevin dynamics. In SIGGRAPH Asia, 2024. 1

    2024

  12. [20]

    C. Jia, Y . Yang, Y . Xia, Y .-T. Chen, Z. Parekh, H. Pham, Q. Le, Y .-H. Sung, Z. Li, and T. Duerig. Scaling up visual and vision-language representation learning with noisy text supervision. In ICML, 2021. 1

    2021

  13. [21]

    Kawano, W

    M. Kawano, W. Kumagai, A. Sannai, Y . Iwasawa, and Y . Matsuo. Group equivariant conditional neural pro- cesses. arXiv preprint arXiv:2102.08759, 2021. 2

    2021 arXiv

  14. [22]

    Keller and Y

    Y . Keller and Y . Shkolnisky. A signal processing ap- proach to symmetry detection. IEEE TIP, 2006. 2

    2006

  15. [23]

    M. U. Khattak, H. Rasheed, M. Maaz, S. Khan, and F. S. Khan. Maple: Multi-modal prompt learning. In CVPR, 2023. 2

    2023

  16. [24]

    Lee and Y

    S. Lee and Y . Liu. Skewed rotation symmetry group detection. IEEE TPAMI, 2009. 2

    2009

  17. [25]

    Lengyel, O

    A. Lengyel, O. Strafforello, R.-J. Bruintjes, A. Gielisse, and J. van Gemert. Color equivariant convolutional networks. In NeurIPS, 2024. 2

    2024

  18. [26]

    Lin, L.-L

    H.-C. Lin, L.-L. Wang, and S.-N. Yang. Extracting periodicity of a regular texture based on autocorrelation functions. Pattern recognition letters, 1997. 2

    1997

  19. [27]

    T.-Y . Lin, P. Goyal, R. Girshick, K. He, and P. Dollár. Focal loss for dense object detection. In ICCV, 2017. 5

    2017

  20. [28]

    I.-J. Liu, R. A. Yeh, and A. G. Schwing. Pic: permuta- tion invariant critic for multi-agent deep reinforcement learning. In Proc. CORL, 2020. 2

    2020

  21. [29]

    I.-J. Liu, Z. Ren, R. A. Yeh, and A. G. Schwing. Se- mantic tracklets: An object-centric representation for visual multi-agent reinforcement learning. In Proc. IROS, 2021. 2

    2021

  22. [30]

    J. Liu, G. Slota, G. Zheng, Z. Wu, M. Park, S. Lee, I. Rauschert, and Y . Liu. Symmetry detection from realworld images competition 2013: Summary and results. In CVPRW, 2013. 1

    2013

  23. [31]

    Y . Liu, H. Hel-Or, C. S. Kaplan, L. Van Gool, et al. Computational symmetry in computer vision and com- puter graphics. Foundations and Trends® in Computer Graphics and Vision, 2010. 1

    2010

  24. [32]

    Lüddecke and A

    T. Lüddecke and A. Ecker. Image segmentation using text and image prompts. In CVPR, 2022. 2

    2022

  25. [33]

    Luká ˇc, D

    M. Luká ˇc, D. S `ykora, K. Sunkavalli, E. Shechtman, O. Jamriška, N. Carr, and T. Pajdla. Nautilus: Re- covering regional symmetry transformations for image editing. ACM TOG, 2017. 1

    2017

  26. [34]

    S. Park, A. Senocak, and J. S. Chung. Can CLIP help sound source localization? In WACV, 2024. 2

    2024

  27. [35]

    H. Pashler. Coordinate frame for symmetry detection and object recognition. Journal of Experimental Psy- chology: Human Perception and Performance, 1990. 1

    1990

  28. [36]

    V . S. N. Prasad and L. S. Davis. Detecting rotational symmetries. In ICCV, 2005. 1, 2

    2005

  29. [37]

    Radford, J

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark, et al. Learning transferable visual models from natural language supervision. In ICML, 2021. 1, 2, 6

    2021

  30. [38]

    M. A. Rahman and R. A. Yeh. Truly scale-equivariant deep nets with fourier layers. In NeurIPS, 2024. 2

    2024

  31. [39]

    M. A. Rahman and R. A. Yeh. Group downsam- pling with equivariant anti-aliasing. In The Thir- teenth International Conference on Learning Repre- sentations, 2025. URL ������������������� ����������������������� . 2

    2025

  32. [40]

    Ramesh, M

    A. Ramesh, M. Pavlov, G. Goh, S. Gray, C. V oss, A. Radford, M. Chen, and I. Sutskever. Zero-shot text-to-image generation. In ICML, 2021. 1

    2021

  33. [41]

    Rasheed, M

    H. Rasheed, M. U. Khattak, M. Maaz, S. Khan, and F. S. Khan. Fine-tuned CLIP models are efficient video learners. In CVPR, 2023. 2

    2023

  34. [42]

    Rauschert, J

    I. Rauschert, J. Liu, K. Brockelhurst, S. Kashyap, and Y . Liu. Symmetry detection competition: A summary of how the competition is carried out. InCVPRW, 2011. 1

    2011

  35. [43]

    Ravanbakhsh, J

    S. Ravanbakhsh, J. Schneider, and B. Poczos. Deep learning with sets and point clouds. In Proc. ICLR workshop, 2017. 2

    2017

  36. [44]

    T. Ren, S. Liu, A. Zeng, J. Lin, K. Li, H. Cao, J. Chen, X. Huang, Y . Chen, F. Yan, et al. Grounded sam: As- sembling open-world models for diverse visual tasks. arXiv preprint arXiv:2401.14159, 2024. 12

    2024 arXiv

  37. [45]

    Rodríguez, A

    I. Rodríguez, A. Gumbert, N. Hempel de Ibarra, J. Kunze, and M. Giurfa. Symmetry is in the eye of the ‘beeholder’: innate preference for bilateral symmetry in flower-naïve bumblebees. Naturwissenschaften, 2004. 1

    2004

  38. [46]

    R. A. Rojas-Gomez, T.-Y . Lim, A. Schwing, M. Do, and R. A. Yeh. Learnable polyphase sampling for shift invariant and equivariant convolutional networks. Advances in Neural Information Processing Systems, 35:35755–35768, 2022. 2

    2022

  39. [47]

    R. A. Rojas-Gomez, T.-Y . Lim, M. N. Do, and R. A. Yeh. Making vision transformers truly shift-equivariant. In CVPR, 2024. 2

    2024

  40. [48]

    Schuhmann, R

    C. Schuhmann, R. Vencu, R. Beaumont, R. Kaczmar- czyk, C. Mullis, A. Katta, T. Coombes, J. Jitsev, and A. Komatsuzaki. Laion-400M: Open dataset of CLIP- filtered 400 million image-text pairs. arXiv preprint arXiv:2111.02114, 2021. 1, 12

    2021 arXiv

  41. [49]

    Seo and M

    A. Seo and M. Cho. Leveraging 3d geometric priors in 2d rotation symmetry detection. In CVPR, 2025. 1, 17

    2025

  42. [50]

    A. Seo, W. Shim, and M. Cho. Learning to discover reflection symmetry via polar matching convolution. In ICCV, 2021. 1, 2, 5, 6, 7, 8, 14, 15, 17

    2021

  43. [51]

    A. Seo, B. Kim, S. Kwak, and M. Cho. Reflection and rotation symmetry detection via equivariant learning. In CVPR, 2022. 1, 2, 3, 5, 6, 7, 8, 13, 14, 15, 17

    2022

  44. [52]

    D. Shen, H. H. Ip, and E. K. Teoh. Robust detection of skewed symmetries by combining local and semi-local affine invariants. Pattern Recognition, 2001. 1, 2

    2001

  45. [53]

    S. N. Sinha, K. Ramnath, and R. Szeliski. Detecting and reconstructing 3d mirror symmetric objects. In ECCV, 2012. 1

    2012

  46. [54]

    Sosnovik, M

    I. Sosnovik, M. Szmaja, and A. Smeulders. Scale- equivariant steerable networks. arXiv preprint arXiv:1910.11093, 2019. 2

    1910 arXiv

  47. [55]

    Tsogkas and I

    S. Tsogkas and I. Kokkinos. Learning-based symmetry detection in natural images. In ECCV, 2012. 2

    2012

  48. [56]

    C. W. Tyler. Human symmetry perception and its com- putational analysis. Psychology Press, 2003. 1

    2003

  49. [57]

    Vetter, T

    T. Vetter, T. Poggio, and H. Bülthoff. The importance of symmetry and virtual views in three-dimensional object recognition. Current Biology, 1994. 1

    1994

  50. [58]

    von Fersen, C

    L. von Fersen, C. S. Manos, B. Goldowsky, and H. Roit- blat. Dolphin detection and conceptualization of sym- metry. Marine mammal sensory systems, 1992. 1

    1992

  51. [59]

    Z. Wang, L. Fu, and Y . Li. Unified detection of skewed rotation, reflection and translation symmetries from affine invariant contour features. Pattern recognition,

  52. [60]

    Z. Wang, Z. Tang, and X. Zhang. Reflection symmetry detection using locally affine invariant edge correspon- dence. IEEE TIP, 2015. 2

    2015

  53. [61]

    Weiler and G

    M. Weiler and G. Cesa. General E(2)-equivariant steer- able cnns. In NeurIPS, 2019. 2

    2019

  54. [62]

    Weiler, M

    M. Weiler, M. Geiger, M. Welling, W. Boomsma, and T. S. Cohen. 3D steerable CNNs: Learning rotationally equivariant features in volumetric data. In NeurIPS, 2018

    2018

  55. [63]

    Weiler, F

    M. Weiler, F. A. Hamprecht, and M. Storath. Learning steerable filters for rotation equivariant cnns. In CVPR,

  56. [64]

    Wilder, M

    J. Wilder, M. Rezanejad, S. Dickinson, K. Siddiqi, A. Jepson, and D. B. Walther. Local contour symmetry facilitates scene categorization. Cognition, 2019. 1

    2019

  57. [65]

    Worrall and M

    D. Worrall and M. Welling. Deep scale-spaces: Equiv- ariance over scale. In NeurIPS, 2019. 2

    2019

  58. [66]

    Z. Wu, Y . Liu, H. Dong, X. Tang, J. Yang, B. Jin, M. Chen, and X. Wei. R2det: Exploring relaxed rota- tion equivariance in 2d object detection. In ICLR, 2025. 1

    2025

  59. [67]

    H. Xu, S. Xie, X. E. Tan, P.-Y . Huang, R. Howes, V . Sharma, S.-W. Li, G. Ghosh, L. Zettlemoyer, and C. Feichtenhofer. Demystifying CLIP data. arXiv preprint arXiv:2309.16671, 2023. 7

    2023 arXiv

  60. [68]

    R. Xu, K. Yang, K. Liu, and F. He. E(2)-equivariant vision transformer. In Proc. UAI, 2023. 2

    2023

  61. [69]

    C. Xue, W. Zhang, Y . Hao, S. Lu, P. H. Torr, and S. Bai. Language matters: A weakly supervised vision- language pre-training approach for scene text detection and spotting. In ECCV, 2022. 2

    2022

  62. [70]

    Yeh, Y .-T

    R. Yeh, Y .-T. Hu, and A. Schwing. Chirality nets for hu- man pose regression. Advances in Neural Information Processing Systems, 32, 2019. 2

    2019

  63. [71]

    R. A. Yeh, A. G. Schwing, J. Huang, and K. Murphy. Diverse generation for multi-agent sports games. In Proc. CVPR, 2019. 2

    2019

  64. [72]

    R. A. Yeh, Y .-T. Hu, M. Hasegawa-Johnson, and A. Schwing. Equivariance discovery by learned parameter-sharing. In International Conference on Artificial Intelligence and Statistics, pages 1527–1545. PMLR, 2022. 2

    2022

  65. [73]

    L. Yuan, D. Chen, Y .-L. Chen, N. Codella, X. Dai, J. Gao, H. Hu, X. Huang, B. Li, C. Li, et al. Florence: A new foundation model for computer vision. arXiv preprint arXiv:2111.11432, 2021. 1

    2021 arXiv

  66. [74]

    C. Yun, S. Bhojanapalli, A. S. Rawat, S. J. Reddi, and S. Kumar. Are transformers universal approximators of sequence-to-sequence functions? In ICLR, 2020. 4

    2020

  67. [75]

    Zaheer, S

    M. Zaheer, S. Kottur, S. Ravanbakhsh, B. Poczos, R. R. Salakhutdinov, and A. J. Smola. Deep sets. InNeurIPS,

  68. [76]

    X. Zhai, B. Mustafa, A. Kolesnikov, and L. Beyer. Sigmoid loss for language image pre-training. InICCV,

  69. [77]

    K. Zhou, J. Yang, C. C. Loy, and Z. Liu. Conditional prompt learning for vision-language models. In CVPR,

  70. [78]

    K. Zhou, J. Yang, C. C. Loy, and Z. Liu. Learning to prompt for vision-language models. IJCV, 2022. 2 Ground Truth PMCNet [50] EquiSym [51] CLIPSym (a) Reflection detection results on DENDI-ref. Ground Truth EquiSym [51] CLIPSym (b) Rotation detection results on DENDI-rot. Fig...

    2022

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.