Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Interactive Eye Aberration Correction for Holographic Near-Eye Display - Augmented Humans 2021 (Oral presentation by Kenta Yamamoto)

Interactive Eye Aberration Correction for Holographic Near-Eye Display - Augmented Humans 2021 (Oral presentation by Kenta Yamamoto)

This slide was presented in Session 6 "Augmented Vision" at the Augmented Humans (AHs) International Conference 2021.
https://augmented-humans.org/

【Publication】
Kenta Yamamoto, Ippei Suzuki, Kosaku Namikawa, Kaisei Sato, and Yoichi Ochiai. 2021 Interactive Eye Aberration Correction for Holographic Near-Eye Display. In Proceedings of the Augmented Humans (AHs) International Conference 2021. (to appear)
https://digitalnature.slis.tsukuba.ac.jp/2021/02/interactive-eye-aberration-correction-for-hned_ahs2021/

【Project page】
https://digitalnature.slis.tsukuba.ac.jp/2021/02/interactive-eye-aberration-correction-for-hned/

【Project movie】
https://www.youtube.com/watch?v=lVYIwx6J4Rg

【Presenter】
Kenta Yamamoto (山本健太)
University of Tsukuba
Graduate School of Library, Information and Media Studies
Digital Nature Group (Yoichi Ochiai)

【Abstract】
Distortions of observed images have been a long-standing problem in near-eye displays. Although many correction methods for optical system-dependent aberrations have been proposed, the image distortions caused by eye aberrations have not been studied thoroughly. In addition to the problem, eye aberrations are individual specific. Therefore, a system capable of correcting the aberration irrespective of the individual is necessary. In this study, we propose an aberration-correctable holographic near-eye display (HNED) that can be used to interactively compensate for image distortions caused by eye aberrations. We formulate a propagation equation that includes eye aberrations in the HNED and developed a GUI that enables a user to correct eye aberrations on their own. In this system, the image displayed on the HNED is updated based on the correction coefficients specified by the user. We performed experiments on human subjects to verify the effectiveness of the proposed method. Our results indicate that the minimum identifiable size in our HNED can be reduced by the aberration correction using our interface, and especially our aberration correction method is useful for the visibility of low visual-acuity users.

Digital Nature Group

February 24, 2021
Tweet

More Decks by Digital Nature Group

Other Decks in Research

Transcript

  1. Interactive Eye Aberration Correction
 for Holographic Near-Eye Display Kenta Yamamoto1,

    Ippei Suzuki1, Kosaku Namikawa1, Kaisei Sato1, Yoichi Ochiai1,2 1University of Tsukuba, Digital Nature Group 2Pixie Dust Technologies, Inc.
  2. 2 Overview of This Work Before Aberration Correction Aberration Correction

    using User Interface After Aberration Correction
  3. 4 Distorted Image by Wave Aberration with Wave Aberration Clear

    Image Distorted Image Problems of Near-Eye Displays: Image Distortion, Focus Cues, 3D Image, Parallax, Occlusion, ... etc.
  4. 6 Previous Work [Maimone et al. 2017] [Takaki et al.

    2018] [Kim et al. 2019, 2020] [Ours. 2021] No GUI with GUI
  5. 7 Previous Work [Maimone et al. 2017] [Takaki et al.

    2018] [Kim et al. 2019, 2020] [Ours. 2021] No GUI with GUI
  6. 8 Previous Work [Maimone et al. 2017] [Takaki et al.

    2018] [Kim et al. 2019, 2020] [Ours. 2021] No GUI with GUI
  7. 9 Previous Work [Maimone et al. 2017] [Takaki et al.

    2018] [Kim et al. 2019, 2020] [Ours. 2021] No GUI with GUI
  8. 15 Aberration Correction Formulation 𝑢𝑠𝑙𝑚 = 𝑃𝑟𝑜𝑝𝑎 (𝑢𝑜𝑏𝑗𝑒𝑐𝑡 ) ∗

    𝑡(𝑥,𝑦) ∗ P𝑟𝑒𝑣 (𝑥,𝑦) propagation from uobject to uslm lens phase distribution general pupil function with opposite aberration
  9. 16 Aberration Correction Simulation 7FSUJDBM"TUJHNBUJTN XJUI$PNQFOTBUJPO 0CMJRVF"TUJHNBUJTN XJUIPVU$PNQFOTBUJPO 0CMJRVF"TUJHNBUJTN XJUI$PNQFOTBUJPO

    7FSUJDBM"TUJHNBUJTN XJUI$PNQFOTBUJPO 0CMJRVF"TUJHNBUJTN XJUIPVU$PNQFOTBUJPO 0CMJRVF"TUJHNBUJTN XJUI$PNQFOTBUJPO uslm ureal uretina Aberration Correction Distorted Corrected Opposite Distortion
  10. 20 User Interface for Aberration Correction Oblique Astigmatism Hyperopia or

    Myopia Vertical or Horizontal Astigmatism Detail of User Interface User Interface with Optical Setup
  11. 21 Workflow of Aberration Correction B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT

    C $IFDLJOH*NBHF *OUFSGBDF Checking Displayed Image B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT C $IFDLJOH*NBHF
  12. 22 Workflow of Aberration Correction B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT

    C $IFDLJOH*NBHF *OUFSGBDF B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT C $IFDLJOH*NBHF *OUFSGBDF B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT C $IFDLJOH*NBHF Adjusting Parameters for Aberration Correction
  13. 23 Workflow of Aberration Correction B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT

    C $IFDLJOH*NBHF *OUFSGBDF B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT C $IFDLJOH*NBHF Checking Displayed Image
  14. 24 Workflow of Aberration Correction B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT

    C $IFDLJOH*NBHF *OUFSGBDF B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT C $IFDLJOH*NBHF *OUFSGBDF B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT C $IFDLJOH*NBHF Adjusting Parameters for Aberration Correction
  15. 25 Workflow of Aberration Correction B "EKVTUJOH1BSBNFUFST 1IBTF *OUFSGBDF %JTQMBZ3FTVMUT

    C $IFDLJOH*NBHF *OUFSGBDF Until the user can get clear image, the user repeats this cycle.
  16. 26 User Experiment Minimum identifiable gap size is measured before

    and after the aberration correction. 18 people participate this user experiment. Gap size
  17. 27 Result of User Experiment Condition 1 Condition 2 Condition

    3 Condition 4 Not Improved Improved Improved Improved
  18. 28 Summary of Contributions Aberration Correctable Holographic Near-Eye Display User

    Interface for Aberration Correction Adjustment User Study
  19. Interactive Eye Aberration Correction
 for Holographic Near-Eye Display Kenta Yamamoto1,

    Ippei Suzuki1, Kosaku Namikawa1, Kaisei Sato1, Yoichi Ochiai1,2 1University of Tsukuba, Digital Nature Group 2Pixie Dust Technologies, Inc.