“All in-Focus Light Field Viewer” by Takahashi and Naemura

  • ©Keita Takahashi and Takeshi Naemura

  • ©Keita Takahashi and Takeshi Naemura

  • ©Keita Takahashi and Takeshi Naemura

Conference:


Type:


Entry Number: 048

Title:

    All in-Focus Light Field Viewer

Presenter(s)/Author(s):



Abstract:


    Light field rendering [Levoy and Hanrahan 1996] is a method for generating free-viewpoint images from pre-acquired images captured by an array of cameras. Since it is based on the constant depth assumption, we can see some focus-like effects on the synthesized images [Isaksen et al. 2000] when the distance between cameras is not small enough. This means that, when the light field is undersampled, objects apart from the assumed depth cannot be synthesized clearly. These objects are out-of-focus, and appear with blurring and ghosting (See Fig. 1 as examples).
    In order to solve this problem, we presented a basic theory in [Takahashi et al. 2003], and will publish its quantitative evaluations in [Takahashi et al. 2004]. In this paper, we developed an interactive viewer based on this theory that can generate all in-focus images at arbitrary viewpoints from undersampled light fields. On the viewer, we introduced a smoothing function to suppress noises which are visible when the viewpoint is moving. In addition, we achieved interactive frame-rates by adopting just two reconstruction filters while we used three in [Takahashi et al. 2003].

References:


    1. Isaksen, A., McMillan, L., and Gortler, S. J. 2000. Dynamically reparameterized light fields. In Proceedings of SIGGRAPH 2000, 297–306.
    2. Levoy, M., and Hanrahan, P. 1996. Light field rendering. In Proceedings of SIGGRAPH 1996, 31–42.
    3. Takahashi, K., Kubota, A., and Naemura, T. 2003. All in-focus view synthesis from under-sampled light fields. In Proceedings of ICAT 2003, 249–256.
    4. Takahashi, K., Kubota, A., and Naemura, T. 2004. A focus measure for light field rendering. To appear in Proceedings of IEEE ICIP 2004.


Additional Images:

©Keita Takahashi and Takeshi Naemura ©Keita Takahashi and Takeshi Naemura ©Keita Takahashi and Takeshi Naemura

Acknowledgements:


    Thanks to Prof. H. Harashima of the University of Tokyo for his helpful discussions.


PDF:



ACM Digital Library Publication:



Overview Page: