Hajime Mihara1, Takuya Funatomi1, Kenichiro Tanaka2,1, Hiroyuki Kubo1, Hajime Nagahara3, Yasuhiro Mukaigawa1
1 Nara Institute of Science and Technology (NAIST), Japan, 2 Osaka University, Japan, 3 Kyushu University, Japan

Appeared in ICCP '16 oral session.

Abstract

In this paper, we describe a supervised four-dimensional (4D) light field segmentation method that uses a graph-cut algorithm. Since 4D light field data has implicit depth information and contains redundancy, it differs from simple 4D hyper-volume. In order to preserve redundancy, we define two neighboring ray types (spatial and angular) in light field data. To obtain higher segmentation accuracy, we also design a learning-based likelihood, called objectness, which utilizes appearance and disparity cues. We show the effectiveness of our method via numerical evaluation and some light field editing applications using both synthetic and real-world light fields.

Video

Presentation

[PowerPoint(.pptx), 26MB]
Notice: We recommend to get original PowerPoint file rather than Slideshare
because PowerPoint file contains many movie contents.
mihara_iccp16_presentation from Hajime Mihara

Publication

  • H. Mihara, T. Funatomi, K. Tanaka, H. Kubo, H. Nagahara, Y. Mukaigawa,
    4D Light-field Segmentation with Spatial and Angular Consistencies,
    In Proceedings of IEEE International Conference on Computational Photography (ICCP), pp. 54-61, 2016.
  • IEEE Xplore (TBA)
  • Preprint (PDF)