Robust super-resolution method for depth images based on image colorization

• Project/Area Number: 15K00246
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: Kogakuin University
• Principal Investigator: KONISHI KATSUMI 工学院大学, 情報学部(情報工学部), 准教授 (20339138)
• Co-Investigator(Kenkyū-buntansha): 古川 利博 東京理科大学, 工学部情報工学科, 教授 (00190140)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Project Status: Completed (Fiscal Year 2017)
• Budget Amount: ¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000); Fiscal Year 2017: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2016: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2015: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
• Keywords: 画像復元 / 画像修復 / 最適化 / 深度画像 / カラリゼーション / 画像処理

Outline of Final Research Achievements

This work proposed a new depth image recovery method based on image colorization. Assuming that neighbor pixels of a depth image have almost the same values when neighbor pixels of a chrominance image have almost the same values, the depth image recovery problem was formulated as the mixed l_1 and l_2 minimization problem and provides an iterative algorithm. Numerical examples show the efficiency of the proposed algorithm.

Research Products

• [Presentation] Representative Pixels Compression Algorithm Using Graph Signal Processing for Colorization-Based Image Coding (2017)
  • Author(s): K. Uruma, K. Saito, T. Takahashi, K. Konishi and T. Furukawa
  • Organizer: IEEE International Conference on Image Processing (ICIP)
  • Int'l Joint Research
• [Presentation] Depth Image Coding Algorithm Via the Colorization Based Image Coding (2016)
  • Author(s): K. Uruma, K. Konishi, T. Takahashi and T. Furukawa
  • Organizer: IEEE 59th International Midwest Symposium on Circuits and Systems
  • Place of Presentation: Abu Dhabi, United Arab Emirates
  • Year and Date: 2016-10-18
  • Int'l Joint Research
• [Presentation] High resolution depth image recovery algorithm based on the modeling of the sum of an average distance image and a surface image (2016)
  • Author(s): K. Uruma, K. Konishi, T. Takahashi and T. Furukawa
  • Organizer: EEE International Conference on Image Processing
  • Place of Presentation: Phoenix, USA
  • Year and Date: 2016-09-25
• [Presentation] 復元誤差のL1ノルム最小化に基づくカラー画像を用いた深度画像符号化手法の提案 (2016)
  • Author(s): 雨車和憲，小西克巳，高橋智博，古川利博
  • Organizer: 電子情報通信学会ソサイエティ大会
  • Place of Presentation: 北海道札幌市
  • Year and Date: 2016-09-22
• [Presentation] スーパーピクセル分割を用いたカラリゼーション符号化に基づく深度画像符号化手法の提案 (2016)
  • Author(s): 雨車和憲，小西克巳，高橋智博，古川利博
  • Organizer: 第１５回情報科学技術フォーラム
  • Place of Presentation: 富山県富山市
  • Year and Date: 2016-09-07
• [Presentation] RGB画像から得られる小平面の張り合わせによる深度画像の超解像手法の提案 (2015)
  • Author(s): 雨車、小西、高橋、古川
  • Organizer: 第30回信号処理シンポジウム
  • Place of Presentation: 福島県いわき市
  • Year and Date: 2015-11-04
• [Presentation] Depth Image Recovery Algorithm using Image Segmentation and Sparse Optimization (2015)
  • Author(s): K. Uruma, K. Konishi, T. Takahashi and T. Furukawa
  • Organizer: International Conference on Simulation Technology
  • Place of Presentation: 富山
  • Year and Date: 2015-10-12
  • Int'l Joint Research
• [Presentation] High resolution depth image recovery algorithm using grayscale image (2015)
  • Author(s): Kazunori Uruma, Katsumi Konishi, Tomohiro Takahashi, Toshihiro Furukawa
  • Organizer: 23rd European Signal Processing Conference (EUSIPCO)
  • Place of Presentation: フランス ニース
  • Year and Date: 2015-08-31
  • Int'l Joint Research