Pith. sign in

REVIEW 3 major objections 4 minor 69 references

Static or Temporal? Semantic Scene Simplification to Aid Wayfinding in Immersive Simulations of Bionic Vision

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Semantic scene simplification—static or time-sequenced—improves wayfinding in simulated bionic vision, with static edges lifting success odds and temporally cycled edges cutting collisions.

desk verdict A genuinely new encoding strategy and a well-grounded simulation, but the Control-first design means the headline baseline comparisons may be partly an order artifact. read the letter →

arxiv 2507.10813 v1 pith:BGOFMQ4J submitted 2025-07-14 cs.HC

Justin M. Kasowski , Apurv Varshney , Michael Beyeler This is my paper
classification cs.HC
keywords bionicvisionsimulatedprostheticsemanticsegmentationtemporalmultiplexingwayfindingvirtualrealityscenesimplificationphosphenerendering
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 tries to establish that a bionic-vision system can help users navigate better by simplifying scenes according to meaning rather than raw edges. In an immersive virtual-reality simulation of prosthetic vision, 18 sighted participants walked across a cluttered town square under three display strategies: an edge-detection baseline, a static overlay of task-relevant semantic edges, and a new temporal version that cycles bicycles, pedestrians, and structures at 200 milliseconds per class. The authors report that both semantic strategies improved performance and user experience relative to the baseline, with distinct trade-offs: the static overlay increased the odds of reaching the goal, while the temporal version increased the odds of finishing without any collision. If the result holds, the limited stimulation budget of retinal implants can be spent on task meaning rather than raw structure, and time-sharing semantic layers is a viable encoding for low-bandwidth visual interfaces.

What carries the argument

The machinery is a content-aware raster: instead of sweeping a fixed spatial pattern over the electrode array, the system cycles through semantic groups in time, with object classes ordered by task relevance and each class given a 200 ms slot—bicycles, then pedestrians, then structural edges. It rides on a psychophysically grounded simulated-prosthetic-vision pipeline: a 10 by 10 epiretinal electrode array rendered with gaze-contingent updating, axon-map spatial distortion, temporal fading and persistence from coupled leaky integrators, and a checkerboard raster that keeps simultaneous activation below safety limits. SemanticRaster repurposes the same temporal budget that safety limits already impose on stimulation, so time-division carries task semantics at no extra bandwidth cost.

What would settle it

Reverse the design: give a fresh group of participants the SemanticEdges or SemanticRaster block first and the Control block last, with the same trial-index covariate. If the success-odds advantage of SemanticEdges (1.84×) and the collision-free advantage of SemanticRaster (2.1×) shrink to null when Control is no longer the first block, the central comparison was confounded by practice rather than by scene simplification.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that semantic scene simplification helps in simulated prosthetic vision, and that static and temporal versions help in different ways. SemanticEdges shows all task-relevant object classes simultaneously and raised the odds of completing the wayfinding task by a factor of 1.84 relative to the Sobel-edge Control; SemanticRaster, which cycles those classes over time with 200 ms slots, did not significantly change success odds but raised the odds of a collision-free run by about 2.1 and cut total collisions by 26 percent, versus 21 percent for SemanticEdges. Neither strategy slowed participants down, the improvements were stable across trials, and both were rated one and a half to two points less difficult than Control. The authors read this as evidence that a static overlay supports global awareness while temporal sequencing reduces clutter, so the better encoding depends on whether the costlier failure is missing the goal or hitting a hazard.

Load-bearing premise

The load-bearing premise is that the Control condition, which every participant experienced first, can be fairly compared with the two semantic conditions after controlling for a linear trial-index covariate; if in-session learning is not captured by that covariate, the claimed benefits of the semantic strategies would be partly an artifact of practice.

Editorial extensions

If this is right

  • A static overlay of all task-relevant semantic classes supports global awareness and increases the odds of completing a wayfinding task, compared with an edge-detection baseline.
  • Time-sequencing those classes increases the odds of finishing without collisions and reduces the total number of collisions, with no cost to completion time.
  • Neither semantic strategy harms performance relative to the baseline, and participants rate both as less difficult than the baseline.
  • The temporal budget already imposed by safe stimulation can carry task semantics by sweeping semantic layers instead of spatial strips, which is directly relevant to raster-based implants.
  • The choice between static and temporal simplification can be guided by the task's failure mode: missing the goal versus hitting a hazard.

Reading between the lines

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

  • Editorial extension: the 200 ms slot length and the fixed class ordering could be made continuous controls, so a system could allocate more of the temporal budget to whichever semantic class is currently most hazardous.
  • Editorial extension: because the paper reports no significant reduction in collisions with moving obstacles, a natural next step is to combine temporal semantic cycling with motion or looming cues.
  • Editorial extension: a hybrid that begins each trial with a brief static SemanticEdges frame and then switches to SemanticRaster may capture both the success benefit and the collision benefit in a single encoding.
  • Editorial extension: the same time-division principle generalizes to any bandwidth-limited visual interface where clutter, not resolution, is the bottleneck.
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 / 4 minor

Summary. The paper proposes SemanticRaster, a content-aware raster strategy that time-multiplexes semantic object classes, and compares it against a static semantic edge overlay (SemanticEdges) and a conventional edge-detection baseline (Control) in an immersive VR simulation of prosthetic vision. Eighteen sighted participants performed a wayfinding task with dynamic obstacles under all three conditions. The authors report that both semantic strategies improve success, collision rates, and perceived difficulty relative to the Control baseline, with SemanticEdges increasing odds of success and SemanticRaster increasing odds of collision-free completion. The paper frames the contribution as design guidance for bandwidth-limited bionic vision and XR displays.

Significance. If the results hold, the paper makes a useful methodological and empirical contribution: it extends checkerboard raster work to content-aware temporal multiplexing, uses a biologically grounded gaze-contingent phosphene simulation, and builds on an open-source platform (BionicVisionXR) that supports reproducibility. The study is carefully instrumented, with transparent model specifications and effect sizes. However, the central baseline comparisons are compromised by a systematic order confound: Control was always the first block, and the linear trial-index covariate used to adjust for learning cannot fully remove the resulting practice effect. Because the load-bearing claim is exactly that both semantic strategies outperform the baseline, the significance of the paper depends on how this confound is handled.

major comments (3)
  1. [§4.4.2, §4.5, §5.1] The within-subject design is confounded by presentation order. Section 4.4.2 states that the Control condition was always presented first, while only the two semantic conditions were counterbalanced. Consequently, all Control trials occupy trial indices 1–10 and all semantic trials occupy later indices. The mixed models in Section 4.5 include a single centered TrialIndex covariate (linear, and in some models with a random slope) to control for learning. This adjustment cannot identify the condition effect: at centered TrialIndex=0 (the average trial), no Control observations exist, so the Control baseline is an extrapolation of a linear learning curve fitted to its first-block data. The non-significant Condition×TrialIndex interactions (e.g., χ²(2)=0.17, p=.92 for success) test whether learning slopes differ between conditions, not whether the first-block baseline is depressed by early-session learning. Thus the headline contrasts (SemanticEdges OR=1.84 for success; SemanticRaster OR=2.1 for collision-free completion) could be produced by trial position rather than by display content. The statement in Section 5.2 that the improvements were driven by the strategies themselves, rather than by learning, is not warranted by this design.
  2. [§5.1, §5.2, Abstract] The 'distinct trade-offs' claim in the abstract and Section 5.1 is not directly established by the reported statistical contrasts. For success, SemanticRaster's improvement over Control is non-significant (β=0.27, p=.24), and no SemanticEdges-versus-SemanticRaster contrast for success is reported. For collision-free completions, SemanticEdges' OR=1.8 is non-significant (p=.086). Where a direct semantic-versus-semantic comparison is reported (total collisions, β=0.067, p=.77), the two strategies are indistinguishable. The data are consistent with the claimed pattern, but the claim of complementary benefits requires either explicit direct contrasts or a more cautious interpretation.
  3. [§4.3.1] The specification of the SemanticRaster schedule is internally inconsistent. Section 4.3.1 says the strategy allocates equal temporal slots to each class (200 ms) but 'with more frequent recurrence of higher-ranked categories.' If the schedule simply cycles bicycles, then pedestrians, then structural edges at equal 200 ms slots, all classes recur every 600 ms and no class is more frequent. The authors should specify whether higher-ranked classes are repeated within each cycle, or clarify that priority is expressed only by ordering rather than by recurrence frequency.
minor comments (4)
  1. [Figure 6 caption] The panel labels in the Figure 6 caption are listed out of alphabetical order (F appears before E); please reorder them for clarity.
  2. [§4.3] Implementation details of the semantic segmentation are not provided (e.g., network architecture, training data, operating frequency). Including a reference or a reproduction note would improve reproducibility.
  3. [§5.1] In the collision-free completion paragraph, the sentence introducing the results says 'both SemanticEdges and SemanticRaster increased the likelihood of a clean run,' but SemanticEdges' effect is non-significant (p=.086). Please clarify that only SemanticRaster reached significance.
  4. [§5.2] Effect sizes for the collision models are reported as percentages without confidence intervals; adding confidence intervals would help readers assess precision.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the semantic-vs-baseline results are direct empirical contrasts; self-citations support component simulation choices but do not determine the outcome.

full rationale

The paper contains no derivation that reduces to its own inputs. Its central claims—that SemanticEdges increases the odds of success and SemanticRaster increases the likelihood of collision-free completions—are empirical contrasts estimated from trial-level outcome data with mixed-effects models in Section 5, not quantities forced by construction. The only fitted parameters imported into the simulation are temporal phosphene dynamics, which were fit to previously published data (Section 4.2.2: 'Parameter values (tau_n = 0.2 s, tau_b = 5 s, and alpha = 0.2) were fitted to reproduce temporal fading and persistence effects reported by Subject 5 of Pérez Fornos et al. [46]'), and they are held fixed across all three experimental conditions. The self-citations that do appear, notably [28] (BionicVisionXR) and [30] (checkerboard raster), support the simulation toolchain and a component of the rendering pipeline applied symmetrically to every condition; because the checkerboard raster is common to Control, SemanticEdges, and SemanticRaster, its prior validation cannot by itself produce the observed condition differences. No step invokes a uniqueness theorem, no fitted value is renamed as a prediction, and no outcome is defined in terms of a manipulated variable. The study's most serious internal-validity weakness is procedural rather than circular: the Control condition was always presented first (Section 4.4.2), so the centered TrialIndex covariate may not fully capture nonlinear practice effects, which is a potential order confound affecting causal interpretation. That is a legitimate methodological risk, but it is not an equation-level equivalence between an input and an output. Accordingly, no circular step is present.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The central claim depends on a simulation stack with several chosen constants and a fixed task schedule, but no invented physical entities. The fitted temporal model and hand-picked raster schedule are the main parameters; the core comparison is empirical.

free parameters (4)
  • Temporal model time constants (tau_n, tau_b, alpha) = tau_n=0.2 s, tau_b=5 s, alpha=0.2
    Fitted to reproduce temporal fading and persistence of Subject 5 in Perez Fornos et al. (Section 4.2.2); single-subject calibration, not independently validated across participants.
  • Axon map spatial parameters (rho, lambda) = rho=200 micrometers, lambda=400 micrometers
    Chosen from earlier psychophysical studies to represent typical epiretinal distortions (Section 4.2.1); not varied in this study, but affects simulation realism.
  • SemanticRaster slot duration = 200 ms per class
    Chosen by the authors; only one schedule was tested, so there is no evidence that this timing is optimal or that results generalize to other periods (Section 4.3).
  • Semantic class priority ordering = bicycles > pedestrians > structural edges
    Derived from one blind consultant and an O&M specialist; task-specific and not empirically optimized (Section 4.3.1).
assumptions (3)
  • domain assumption Sighted participants in simulated prosthetic vision provide valid within-subject evidence for comparing encoding strategies in eventual implant users.
    Stated in Sections 1 and 4.1; acknowledged that sighted users cannot model long-term adaptation or device-specific variability.
  • domain assumption The simplified Horsager temporal model with parameters from one subject approximates the temporal dynamics of phosphenes for the whole participant group.
    Section 4.2.2 fits to Subject 5 of Perez Fornos et al.; the paper does not validate the fit across subjects.
  • domain assumption The 10x10 electrode array and 14.6x14.6 degree field of view approximate a current-generation epiretinal implant sufficiently for design comparison.
    Section 4.2 states the array is inspired by Argus II; real implants have wider peripheral fields, and the simulation restricts the visible area.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Static or Temporal? Semantic Scene Simplification to Aid Wayfinding in Immersive Simulations of Bionic Vision." pith.science (2026). https://pith.science/paper/BGOFMQ4J

@misc{pith2026250710813,
  author       = {Pith},
  title        = {Pith review of: Static or Temporal? Semantic Scene Simplification to Aid Wayfinding in Immersive Simulations of Bionic Vision},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BGOFMQ4J}},
  note         = {Machine review of arXiv:2507.10813}
}
read the original abstract

Visual neuroprostheses (bionic eye) aim to restore a rudimentary form of vision by translating camera input into patterns of electrical stimulation. To improve scene understanding under extreme resolution and bandwidth constraints, prior work has explored computer vision techniques such as semantic segmentation and depth estimation. However, presenting all task-relevant information simultaneously can overwhelm users in cluttered environments. We compare two complementary approaches to semantic preprocessing in immersive virtual reality: SemanticEdges, which highlights all relevant objects at once, and SemanticRaster, which staggers object categories over time to reduce visual clutter. Using a biologically grounded simulation of prosthetic vision, 18 sighted participants performed a wayfinding task in a dynamic urban environment across three conditions: edge-based baseline (Control), SemanticEdges, and SemanticRaster. Both semantic strategies improved performance and user experience relative to the baseline, with each offering distinct trade-offs: SemanticEdges increased the odds of success, while SemanticRaster boosted the likelihood of collision-free completions. These findings underscore the value of adaptive semantic preprocessing for prosthetic vision and, more broadly, may inform the design of low-bandwidth visual interfaces in XR that must balance information density, task relevance, and perceptual clarity.

Figures

Figures reproduced from arXiv: 2507.10813 by the authors.

Figure 1
Figure 1. Scene simplification for bionic vision. (A) Bionic vision systems (e.g., retinal implants) capture visual information [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Simplified overview of the SPV pipeline. Unity’s virtual camera captured scenes while tracking gaze position (“Image [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Scene simplification strategies tested in the study. The raw RGB image (top center) was processed using three methods. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: The simulated town square environment, high [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Collision feedback system. When a virtual collision [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Task performance metrics across conditions. (A) Success rate: proportion of trials completed without collisions or [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Difficulty ratings: Participants rated each condi [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 8
Figure 8. Figure 8: Eye tracking accuracy of the HTC Vive Pro. The [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

69 extracted references · 51 canonical work pages

  1. [1]

    Weiland, Ariel Rokem, Geoffrey M

    Michael Beyeler, Devyani Nanduri, James D. Weiland, Ariel Rokem, Geoffrey M. Boynton, and Ione Fine. 2019. A model of ganglion axon pathways accounts for percepts elicited by retinal implants. Scientific Reports 9, 1 (June 2019), 1–16. https://doi.org/10.1038/s41598-019-45416-4

    doi:10.1038/s41598-019-45416-4 2019

  2. [3]

    Michael Beyeler and Melani Sanchez-Garcia. 2022. Towards a Smart Bionic Eye: AI-powered artificial vision for the treatment of incurable blindness. Journal of Neural Engineering 19, 6 (Dec. 2022), 063001. https://doi.org/10.1088/1741- 2552/aca69d Publisher: IOP Publishing

    doi:10.1088/1741- 2022

  3. [4]

    Rupert R. A. Bourne, Jaimie Adelson, Seth Flaxman, Paul Briant, Michele Bottone, Theo Vos, Kovin Naidoo, Tasanee Braithwaite, Maria Cicinelli, Jost Jonas, Hans Limburg, Serge Resnikoff, Alex Silvester, Vinay Nangia, and Hugh R. Taylor

    work page

  4. [5]

    Boyle, Anthony J

    Justin R. Boyle, Anthony J. Maeder, and Wageeh W. Boles. 2008. Region-of-interest processing for electronic visual prostheses. Journal of Electronic Imaging 17, 1 (Jan. 2008), 013002. https://doi.org/10.1117/1.2841708 Publisher: International Society for Optics and Photonics

    work page doi:10.1117/1.2841708 2008

  5. [6]

    Barry, Uday K

    Avi Caspi, Michael P. Barry, Uday K. Patel, Michelle Armenta Salas, Jessy D. Dorn, Arup Roy, Soroush Niketeghad, Robert J. Greenberg, and Nader Pouratian

    work page

  6. [7]

    S. C. Chen, G. J. Suaning, J. W. Morley, and N. H. Lovell. 2009. Simulating prosthetic vision: I. Visual models of phosphenes. Vision Research 49, 12 (June 2009), 1493–506

    work page 2009

  7. [8]

    Roelfsema

    Xing Chen, Feng Wang, Eduardo Fernandez, and Pieter R. Roelfsema. 2020. Shape perception via a high-channel-count neuroprosthesis in monkey visual cortex. Science 370, 6521 (Dec. 2020), 1191–1196. https://doi.org/10.1126/science.abd7435 Publisher: American Association for the Advancement of Science Section: Re- search Article

    doi:10.1126/science.abd7435 2020

  8. [9]

    Daeun Joyce Chung, Muya Guoji, Nina Mindel, Alexis Malkin, Fernando Al- berotrio, Shane Lowe, Chris McNally, Casandra Xavier, and Paul Ruvolo. 2024. Large-scale, Longitudinal, Hybrid Participatory Design Program to Create Nav- igation Technology for the Blind. https://doi.org/10.48550/arXiv.2410.00192 arXiv:2410.00192

    work page Pith review arXiv doi:10.48550/arxiv.2410.00192 2024

Show all 69 references
  1. [10]

    Dagnelie, P

    G. Dagnelie, P. Christopher, A. Arditi, L. da Cruz, J. L. Duncan, A. C. Ho, L. C. de Koo, J. A. Sahel, P. E. Stanga, G. Thumann, Y. Wang, M. Arsiero, J. D. Dorn, R. J. Greenberg, and I. I. Study Group Argus. 2016. Performance of real-world functional vision tasks by blind subj...

    2016 doi

  2. [11]

    Dagnelie, P

    G. Dagnelie, P. Keane, V. Narla, L. Yang, J. Weiland, and M. Humayun. 2007. Real and virtual mobility performance in simulated prosthetic vision. J Neural Eng 4, 1 (March 2007), S92–101. https://doi.org/10.1088/1741-2560/4/1/S11

    2007 doi

  3. [12]

    Cordelia Erickson-Davis and Helma Korzybska. 2021. What do blind people “see” with retinal prostheses? Observations and qualitative reports of epiretinal implant users. PLOS ONE 16, 2 (Feb. 2021), e0229189. https://doi.org/10.1371/ journal.pone.0229189 Publisher: Public Librar...

    2021

  4. [13]

    Eduardo Fernandez. 2018. Development of visual Neuroprostheses: trends and challenges. Bioelectronic Medicine 4, 1 (Aug. 2018), 12. https://doi.org/10.1186/ s42234-018-0013-8

    2018

  5. [14]

    Lozano, Sebastian Peña, Maria Dolores Grima, Alfonso Rodil, Bernardeta Gómez, Xing Chen, Pieter R

    Eduardo Fernández, Arantxa Alfaro, Cristina Soto-Sánchez, Pablo Gonzalez- Lopez, Antonio M. Lozano, Sebastian Peña, Maria Dolores Grima, Alfonso Rodil, Bernardeta Gómez, Xing Chen, Pieter R. Roelfsema, John D. Rolston, Tyler S. Davis, and Richard A. Normann. 2021. Visual perce...

    2021 doi

  6. [15]

    Ione Fine and Geoffrey M. Boynton. 2024. A virtual patient simulation modeling the neural and perceptual effects of human visual cortical stimulation, from pulse trains to percepts. Scientific Reports 14, 1 (July 2024), 17400. https://doi.org/10. 1038/s41598-024-65337-1 Publis...

    2024

  7. [16]

    Bhanuka Gamage, Nicola McDowell, Dijana Kovacic, Leona Holloway, Thanh- Toan Do, Nicholas Price, Arthur Lowery, and Kim Marriott. 2025. Smart Glasses for CVI: Co-Designing Extended Reality Solutions to Support Environmental Perception by People with Cerebral Visual Impairment....

    2025 doi arXiv

  8. [17]

    Duane R Geruschat, Thomas P Richards, Aries Arditi, Lyndon da Cruz, Gislin Dagnelie, Jessy D Dorn, Jacque L Duncan, Allen C Ho, Lisa C Olmos de Koo, José-Alain Sahel, Paulo E Stanga, Gabriele Thumann, Vizhong Wang, and Robert J Greenberg. 2016. An analysis of observer-rated fu...

    2016 doi

  9. [18]

    Jacob Granley and Michael Beyeler. 2021. A Computational Model of Phosphene Appearance for Epiretinal Prostheses. In 2021 43rd Annual International Confer- ence of the IEEE Engineering in Medicine & Biology Society (EMBC) . 4477–4481. https://doi.org/10.1109/EMBC46164.2021.962...

    2021

  10. [19]

    Brian Guenter, Mark Finch, Steven Drucker, Desney Tan, and John Snyder. 2012. Foveated 3D graphics. ACM Trans. Graph. 31, 6 (Nov. 2012), 164:1–164:10. https: //doi.org/10.1145/2366145.2366183

    2012

  11. [20]

    Nicole Han, Sudhanshu Srivastava, Aiwen Xu, Devi Klein, and Michael Beyeler

  12. [21]

    J. S. Hayes, V. T. Yin, D. Piyathaisere, J. D. Weiland, M. S. Humayun, and G. Dagnelie. 2003. Visually guided performance of simple tasks using simulated prosthetic vision. Artif Organs 27, 11 (Nov. 2003), 1016–28

    2003

  13. [22]

    Heidi Ahmed Holiel, Sahar Ali Fawzi, and Walid Al-Atabany. [n. d.]. Pre- processing visual scenes for retinal prosthesis systems: A comprehensive review. Artificial Organs n/a, n/a ([n. d.]). https://doi.org/10.1111/aor.14824 _eprint: https://onlinelibrary.wiley.com/doi/pdf/10...

    doi

  14. [23]

    InAugmented Humans Conference 2021 (AHs’21)

    Deep Learning–Based Scene Simplification for Bionic Vision. InAugmented Humans Conference 2021 (AHs’21) . Association for Computing Machinery, New York, NY, USA, 45–54. https://doi.org/10.1145/3458709.3458982

    2021

  15. [24]

    Horsager, S

    A. Horsager, S. H. Greenwald, J. D. Weiland, M. S. Humayun, R. J. Greenberg, M. J. McMahon, G. M. Boynton, and I. Fine. 2009. Predicting visual sensitivity in retinal prosthesis patients. Invest Ophthalmol Vis Sci 50, 4 (April 2009), 1483–91. https://doi.org/10.1167/iovs.08-2595

    2009 doi

  16. [25]

    Weiland, and Michael Beyeler

    Yuchen Hou, Devyani Nanduri, Jacob Granley, James D. Weiland, and Michael Beyeler. 2024. Axonal stimulation affects the linear summation of single-point perception in three Argus II users.Journal of Neural Engineering 21, 2 (April 2024), 026031. https://doi.org/10.1088/1741-25...

    2024 doi

  17. [26]

    Hoogsteen, Sarit Szpiro, Gabriel Kreiman, and Eli Peli

    Karst M.P. Hoogsteen, Sarit Szpiro, Gabriel Kreiman, and Eli Peli. 2022. Beyond the Cane: Describing Urban Scenes to Blind People for Mobility Tasks. ACM Transactions on Accessible Computing (Feb. 2022). https://doi.org/10.1145/3522757 Just Accepted

    2022 doi

  18. [27]

    Fabbri, Guy Eichler, Zhe Li, Konstantin Willeke, Katie E

    Taesung Jung, Nanyu Zeng, Jason D. Fabbri, Guy Eichler, Zhe Li, Konstantin Willeke, Katie E. Wingel, Agrita Dubey, Rizwan Huq, Mohit Sharma, Yaoxing Hu, Girish Ramakrishnan, Kevin Tien, Paolo Mantovani, Abhinav Parihar, Heyu Yin, Denise Oswalt, Alexander Misdorp, Ilke Uguz, To...

    2024

  19. [28]

    Justin Kasowski and Michael Beyeler. 2022. Immersive Virtual Reality Simula- tions of Bionic Vision. In Augmented Humans 2022 (AHs 2022) . Association for Computing Machinery, New York, NY, USA, 82–93. https://doi.org/10.1145/ 3519391.3522752

    2022

  20. [29]

    Yuchen Hou, Laya Pullela, Jiaxin Su, Sriya Aluru, Shivani Sista, Xiankun Lu, and Michael Beyeler. 2024. Predicting the Temporal Dynamics of Prosthetic Vision. In 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 1–4. http...

    2024

  21. [30]

    Justin M Kasowski, Apurv Varshney, Roksana Sadeghi, and Michael Beyeler

  22. [31]

    Legge and Susana T.L

    Gordon E. Legge and Susana T.L. Chung. 2016. Low Vision and Plasticity: Im- plications for Rehabilitation. Annual Review of Vision Science 2, 1 (Oct. 2016), 321–343. https://doi.org/10.1146/annurev-vision-111815-114344

    2016 doi

  23. [32]

    Johnson, Ryan Neydavood, Anvitha Akkaraju, and Michael Beyeler

    Justin Kasowski, Byron A. Johnson, Ryan Neydavood, Anvitha Akkaraju, and Michael Beyeler. 2023. A systematic review of extended reality (XR) for un- derstanding and augmenting vision loss. Journal of Vision 23, 5 (May 2023), 5. https://doi.org/10.1167/jov.23.5.5

    2023 doi

  24. [33]

    Walker, Viorica Botea, and Adele F

    Paulette Lieby, Nick Barnes, Chris McCarthy, Nianjun Liu, Hugh Dennett, Ja- nine G. Walker, Viorica Botea, and Adele F. Scott. 2011. Substituting depth for intensity and real-time phosphene rendering: visual navigation under low vision conditions. In Annual International Confe...

    2011

  25. [34]

    Lorach, G

    H. Lorach, G. Goetz, R. Smith, X. Lei, Y. Mandel, T. Kamins, K. Mathieson, P. Huie, J. Harris, A. Sher, and D. Palanker. 2015. Photovoltaic restoration of sight with high visual acuity. Nat Med 21, 5 (May 2015), 476–82. https://doi.org/10.1038/nm.3851

    2015 doi

  26. [35]

    Javier Martinez-Alvarez, J

    Antonio Lozano, Juan Sebastian Suarez, Cristina Soto-Sanchez, Javier Garrigos, J. Javier Martinez-Alvarez, J. Manuel Ferrandez, and Eduardo Fernandez. 2020. Neurolight: A Deep Learning Neural Interface for Cortical Visual Prostheses. International Journal of Neural Systems (Ma...

    2020

  27. [36]

    Heng Li, Xiaofan Su, Jing Wang, Han Kan, Tingting Han, Yajie Zeng, and Xinyu Chai. 2018. Image processing strategies based on saliency segmentation for object recognition under simulated prosthetic vision. Artificial Intelligence in Medicine 84 (Jan. 2018), 64–78. https://doi....

    2018 doi

  28. [37]

    Y. H. Luo and L. da Cruz. 2016. The Argus((R)) II Retinal Prosthesis System. Prog Retin Eye Res 50 (Jan. 2016), 89–107. https://doi.org/10.1016/j.preteyeres.2015.09. 003

    2016 doi

  29. [38]

    Matteucci, Alejandro Barriga-Rivera, Calvin D

    Paul B. Matteucci, Alejandro Barriga-Rivera, Calvin D. Eiber, Nigel H. Lovell, John W. Morley, and Gregg J. Suaning. 2016. The Effect of Electric Cross-Talk in Retinal Neurostimulation. Investigative Ophthalmology & Visual Science 57, 3 (March 2016), 1031–1037. https://doi.org...

    2016 doi

  30. [39]

    Walker, Paulette Lieby, Adele Scott, and Nick Barnes

    Chris McCarthy, Janine G. Walker, Paulette Lieby, Adele Scott, and Nick Barnes

  31. [40]

    Wen Lik Dennis Lui, Damien Browne, Lindsay Kleeman, Tom Drummond, and Wai Ho Li. 2012. Transformative Reality: improving bionic vision with robotic sensing. In Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine...

    2012 doi

  32. [41]

    Turkstra, Ebenezer Larnyo, and Michael Beyeler

    Lucas Nadolskis, Lily M. Turkstra, Ebenezer Larnyo, and Michael Beyeler. 2024. Aligning Visual Prosthetic Development With Implantee Needs. Translational Vision Science & Technology 13, 11 (Nov. 2024), 28. https://doi.org/10.1167/tvst. 13.11.28

    2024 doi

  33. [42]

    Daniel Palanker, Yannick Le Mer, Saddek Mohand-Said, Mahiul Muqit, and Jose A. Sahel. 2020. Photovoltaic Restoration of Central Vision in Atrophic Age-Related Macular Degeneration. Ophthalmology (Feb. 2020). https://doi.org/10.1016/j. ophtha.2020.02.024

    2020 doi

  34. [43]

    Nadia Paraskevoudi and John S. Pezaris. 2019. Eye Movement Compensation and Spatial Updating in Visual Prosthetics: Mechanisms, Limitations and Future Directions. Frontiers in Systems Neuroscience 12 (2019). https://doi.org/10.3389/ fnsys.2018.00073

    2019

  35. [44]

    Anjul Patney, Joohwan Kim, Marco Salvi, Anton Kaplanyan, Chris Wyman, Nir Benty, Aaron Lefohn, and David Luebke. 2016. Perceptually-based foveated virtual reality. In ACM SIGGRAPH 2016 Emerging Technologies (SIGGRAPH ’16) . Association for Computing Machinery, New York, NY, US...

    2016

  36. [45]

    Elon Musk and Neuralink. 2019. An Integrated Brain-Machine Interface Platform With Thousands of Channels. Journal of Medical Internet Research 21, 10 (Oct. 2019), e16194. https://doi.org/10.2196/16194 Company: Journal of Medical Internet Research Distributor: Journal of Medica...

    2019 doi

  37. [46]

    Angélica Pérez Fornos, Jörg Sommerhalder, Lyndon da Cruz, Jose Alain Sahel, Sad- dek Mohand-Said, Farhad Hafezi, and Marco Pelizzone. 2012. Temporal Properties of Visual Perception on Electrical Stimulation of the Retina.Investigative Ophthal- mology & Visual Science 53, 6 (20...

    2012 doi

  38. [47]

    Rajesh P. N. Rao. 2020. Brain Co-Processors: Using AI to Restore and Augment Brain Function. (Dec. 2020). https://arxiv.org/abs/2012.03378v1

    2020 arXiv

  39. [48]

    Alex Rasla and Michael Beyeler. 2022. The Relative Importance of Depth Cues and Semantic Edges for Indoor Mobility Using Simulated Prosthetic Vision in Immer- sive Virtual Reality. In Proceedings of the 28th ACM Symposium on Virtual Reality Software and Technology (VRST ’22) ....

    2022

  40. [49]

    Reis, Carla S

    Catarina I. Reis, Carla S. Freire, Joaquin Fernández, and Josep M. Monguet. 2011. Patient Centered Design: Challenges and Lessons Learned from Working with Health Professionals and Schizophrenic Patients in e-Therapy Contexts. In EN- TERprise Information Systems (Communication...

    2011 doi

  41. [50]

    Guerrero

    Alejandro Perez-Yus, Jesus Bermudez-Cameo, Gonzalo Lopez-Nicolas, and Jose J. Guerrero. 2017. Depth and Motion Cues With Phosphene Patterns for Pros- thetic Vision. 1516–1525. http://openaccess.thecvf.com/content_ICCV_2017_ workshops/w22/html/Perez-Yus_Depth_and_Motion_ICCV_20...

    2017

  42. [51]

    Barry, Paul Gibson, Avi Caspi, Arup Roy, Duane R

    Roksana Sadeghi, Arathy Kartha, Michael P. Barry, Paul Gibson, Avi Caspi, Arup Roy, Duane R. Geruschat, and Gislin Dagnelie. 2024. Benefits of thermal and distance-filtered imaging for wayfinding with prosthetic vision. Scientific Reports 14, 1 (Jan. 2024), 1313. https://doi.o...

    2024 doi

  43. [52]

    Guerrero

    Melani Sanchez-Garcia, Ruben Martinez-Cantin, and Josechu J. Guerrero. 2019. Indoor Scenes Understanding for Visual Prosthesis with Fully Convolutional Networks. In VISIGRAPP. https://doi.org/10.5220/0007257602180225

    2019 doi

  44. [53]

    Guerrero

    Melani Sanchez-Garcia, Ruben Martinez-Cantin, and Jose J. Guerrero. 2020. Se- mantic and structural image segmentation for prosthetic vision. PLOS ONE 15, 1 (Jan. 2020), e0227677. https://doi.org/10.1371/journal.pone.0227677

    2020 doi

  45. [54]

    Second Sight. 2013. Argus® II Retinal Prosthesis System Surgeon Manual . Number 900029-001 Rev C. Second Sight Medical Products, Inc., Sylmar, CA. https: //www.accessdata.fda.gov/cdrh_docs/pdf11/h110002c.pdf

    2013

  46. [55]

    J. F. Rizzo, J. Wyatt, J. Loewenstein, S. Kelly, and D. Shire. 2003. Perceptual efficacy of electrical stimulation of human retina with a microelectrode array during short-term surgical trials. Invest Ophthalmol Vis Sci 44, 12 (Dec. 2003), 5362–9

    2003

  47. [56]

    Stingl, K

    K. Stingl, K. U. Bartz-Schmidt, D. Besch, C. K. Chee, C. L. Cottriall, F. Gekeler, M. Groppe, T. L. Jackson, R. E. MacLaren, A. Koitschev, A. Kusnyerik, J. Neffendorf, J. Nemeth, M. A. Naeem, T. Peters, J. D. Ramsden, H. Sachs, A. Simpson, M. S. Singh, B. Wilhelm, D. Wong, and...

    2015 doi

  48. [57]

    H Christiaan Stronks and Gislin Dagnelie. 2014. The functional performance of the Argus II retinal prosthesis. Expert Review of Medical Devices 11, 1 (Jan. 2014), 23–30. https://doi.org/10.1586/17434440.2014.862494 Publisher: Taylor & Francis _eprint: https://doi.org/10.1586/1...

    2014

  49. [58]

    Jacob Thomas Thorn, Naig Aurelia Ludmilla Chenais, Sandrine Hinrichs, Marion Chatelain, and Diego Ghezzi. 2021. Virtual reality validation of naturalistic modulation strategies to counteract fading in retinal stimulation . Technical Report. 2021.11.17.468930 pages. https://doi...

    2021 doi

  50. [59]

    Jacob Thomas Thorn, Enrico Migliorini, and Diego Ghezzi. 2020. Virtual reality simulation of epiretinal stimulation highlights the relevance of the visual angle in prosthetic vision. Journal of Neural Engineering 17, 5 (Nov. 2020), 056019. https://doi.org/10.1088/1741-2552/abb...

    2020 doi

  51. [60]

    Sinclair, Mohit N

    Nicholas C. Sinclair, Mohit N. Shivdasani, Thushara Perera, Lisa N. Gillespie, Hugh J. McDermott, Lauren N. Ayton, and Peter J. Blamey. 2016. The Appearance of Phosphenes Elicited Using a Suprachoroidal Retinal Prosthesis. Investigative Ophthalmology & Visual Science 57, 11 (S...

    2016 doi

  52. [61]

    Philip R. Troyk. 2017. The Intracortical Visual Prosthesis Project. In Artificial Vision: A Clinical Guide, Veit Peter Gabel (Ed.). Springer International Publishing, Cham, 203–214. https://doi.org/10.1007/978-3-319-41876-6_16

    2017 doi

  53. [62]

    Macé, and Christophe Jouffrais

    Victor Vergnieux, Marc J.-M. Macé, and Christophe Jouffrais. 2017. Simplifica- tion of Visual Rendering in Simulated Prosthetic Vision Facilitates Navigation. Artificial Organs 41, 9 (Sept. 2017), 852–861. https://doi.org/10.1111/aor.12868 Publisher: John Wiley & Sons, Ltd

    2017 doi

  54. [63]

    Weiland, Steven T

    James D. Weiland, Steven T. Walston, and Mark S. Humayun. 2016. Electrical Stimulation of the Retina to Produce Artificial Vision. Annual Review of Vision Science 2, 1 (2016), 273–294. https://doi.org/10.1146/annurev-vision-111815- 114425

    2016 doi

  55. [64]

    R. G. H. Wilke, G. Khalili Moghadam, N. H. Lovell, G. J. Suaning, and S. Dokos

  56. [65]

    Titchener, David A

    Samuel A. Titchener, David A. X. Nayagam, Jessica Kvansakul, Maria Kolic, Eliz- abeth K. Baglin, Carla J. Abbott, Myra B. McGuinness, Lauren N. Ayton, Chi D. Luu, Steven Greenstein, William G. Kentler, Mohit N. Shivdasani, Penelope J. Allen, and Matthew A. Petoe. 2022. A Secon...

    2022 doi

  57. [2011]

    Journal of Neural Engineering 8, 4 (June 2011), 046016

    Electric crosstalk impairs spatial resolution of multi-electrode arrays in retinal implants. Journal of Neural Engineering 8, 4 (June 2011), 046016. https: //doi.org/10.1088/1741-2560/8/4/046016 Static or Temporal? Semantic Scene Simplification to Aid Wayfinding in Immersive S...

    2011 doi

  58. [2014]

    Journal of Neural Engineering 12, 1 (Nov

    Mobility and low contrast trip hazard avoidance using augmented depth. Journal of Neural Engineering 12, 1 (Nov. 2014), 016003. https://doi.org/10.1088/ 1741-2560/12/1/016003 Publisher: IOP Publishing

    2014

  59. [2020]

    Investigative Ophthalmology & Visual Science 61, 7 (June 2020), 2317–2317

    Global Prevalence of Blindness and Distance and Near Vision Impairment in 2020: progress towards the Vision 2020 targets and what the future holds. Investigative Ophthalmology & Visual Science 61, 7 (June 2020), 2317–2317. https: //iovs.arvojournals.org/article.aspx?articleid=2767477

    2020

  60. [2021]

    Brain Stimulation 14, 4 (July 2021), 851–860

    Eye movements and the perceived location of phosphenes generated by intracranial primary visual cortex stimulation in the blind. Brain Stimulation 14, 4 (July 2021), 851–860. https://doi.org/10.1016/j.brs.2021.04.019

    2021 doi

  61. [2025]

    Journal of Neural Engineering (2025)

    Simulated prosthetic vision confirms checkerboard as an effective raster pattern for epiretinal implants. Journal of Neural Engineering (2025). https: //doi.org/10.1088/1741-2552/adecc4

    2025 doi

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.