A study on efficient light field compression using multi-focus images

2013 Fiscal Year Final Research Report

• Project/Area Number: 23700224
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: National Institute of Informatics
• Principal Investigator: KODAMA Kazuya 国立情報学研究所, コンテンツ科学研究系, 准教授 (80321579)
• Project Period (FY): 2011 – 2013
• Keywords: 映像メディア / 画像符号化 / ３次元画像

Research Abstract

For advanced three-dimensional visual media, arrays of multiple cameras are used to capture light fields from scenes. However, such multi-dimensional light fields are composed of enormous data. In this research, a method for efficient light field compression is proposed using three-dimensional multi-focus images synthesized from four-dimensional light fields. The method gains good quality over 30dB even when the compression ratio is 1/1000, where the conventional method using disparity compensation suffers from awful degradation of reconstructed images. The research papers received two Best Paper Awards and have been published in the top conferences and journals of image processing. The proposed method will be extended to deal with not only light fields composed of multi-view images but also light field video.

Research Products

• [Journal Article] Efficient Reconstruction of All-in-Focus Images Through Shifted Pinholes from Multi-Focus Images for Dense Light Field Synthesis and Rendering (2013)
  • Author(s): Kazuya Kodama and Akira Kubota
  • Journal Title: IEEE Transactions on Image Processing Volume: 22(11) Pages: 4407-4421
  • DOI: 10.1109/TIP.2013.2273668
  • Peer Reviewed
• [Journal Article] 焦点ぼけ構造を介した光線空間圧縮の基礎検討 (2012)
  • Author(s): 坂元 敬, 児玉 和也, 浜本 隆之
  • Journal Title: 電子情報通信学会論文誌J Volume: 95-D(9) Pages: 1678-1681
  • Peer Reviewed
• [Presentation] Linear View/Image Restoration for Dense Light Fields (2014)
  • Author(s): Kazuya KODAMA and Akira KUBOTA
  • Organizer: IEEE 2014 International Conference on Image Processing (ICIP 2014)
  • Place of Presentation: France (Accepted)
  • Year and Date: 20140000
• [Presentation] Free viewpoint image reconstruction by using linear filters for multi-focus images on FPGA (2013)
  • Author(s): Yusuke Minato, Kazuya Kodama, and Takayuki Hamamoto
  • Organizer: International Conference on Three Dimensional Systems and Applications 2013 (3DSA 2013), 1-6
  • Place of Presentation: Japan
  • Year and Date: 2013-06-28
• [Presentation] A STUDY ON EFFICIENT COMPRESSION OF MULTI-FOCUS IMAGES FOR DENSE LIGHT-FIELD RECONSTRUCTION (2012)
  • Author(s): Takashi SAKAMOTO, Kazuya KODAMA, and Takayuki HAMAMOTO
  • Organizer: IEEE Visual Communications and Image Processing 2012 (VCIP 2012),FP.150 1-6
  • Place of Presentation: USA
  • Year and Date: 2012-11-30
• [Presentation] 焦点ぼけ画像群を介した光線空間圧縮における3次元SPIHT の適用 (2012)
  • Author(s): 坂元 敬, 児玉 和也, 浜本 隆之
  • Organizer: 画像符号化シンポジウム(PCSJ2012), P-3.02 39-40
  • Place of Presentation: 日本(優秀論文賞 受賞)
  • Year and Date: 2012-10-25
• [Presentation] A NOVEL SCHEME FOR 4-D LIGHT-FIELD COMPRESSION BASED ON 3-D REPRESENTATION BY MULTI-FOCUS IMAGES (2012)
  • Author(s): Takashi SAKAMOTO, Kazuya KODAMA, and Takayuki HAMAMOTO
  • Organizer: IEEE 2012 International Conference on Image Processing (ICIP 2012), 2901-2904
  • Place of Presentation: USA
  • Year and Date: 2012-10-03
• [Presentation] Effective Global Prediction for Dense Light-Field Compression by Using Synthesized Multi-focus Images (2012)
  • Author(s): Takashi SAKAMOTO, Kazuya KODAMA, and Takayuki HAMAMOTO
  • Organizer: ACM The 39th International Conference and Exhibition on Computer Graphics and Interactive Techniques (SIGGRAPH 2012), Posters No.53
  • Place of Presentation: USA
  • Year and Date: 2012-08-07
• [Presentation] 焦点ぼけ画像群における周波数の偏りを考慮した光線空間の高能率圧縮の検討 (2011)
  • Author(s): 坂元 敬, 児玉 和也, 浜本 隆之
  • Organizer: 3次元画像コンファレンス2011, 126-129
  • Place of Presentation: 日本(優秀論文賞 受賞)
  • Year and Date: 2011-07-15
• [Remarks] ① 焦点ぼけ画像群を介した光線空間情報の高能率圧縮, 児玉 和也, 超臨場感コミュニケーション産学官フォーラム(URCF)「超臨場感立体映像技術に関する調査WGセミナー3 立体映像の最新画像・信号処理技術」, 2013年1月31日