Drawing Assistance for Creating Animation via Video Database

• Project/Area Number: 16H07411
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Single-year Grants
• Research Field: Perceptual information processing
• Research Institution: National Institute of Informatics
• Principal Investigator: Matsui Yusuke 国立情報学研究所, コンテンツ科学研究系, 特任研究員 (80780676)
• Project Period (FY): 2016-08-26 – 2018-03-31
• Project Status: Completed (Fiscal Year 2017)
• Budget Amount: ¥2,990,000 (Direct Cost: ¥2,300,000、Indirect Cost: ¥690,000); Fiscal Year 2017: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000); Fiscal Year 2016: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
• Keywords: 近似最近傍探索 / 画像検索 / コンピュータアニメーション / 描画支援 / 映像検索

Outline of Final Research Achievements

We conducted a research on assisting contents-creation via image search with approximate nearest neighbor search. For a theoretical contribution, we proposed a data structure with product-quantization codes (PQ codes) and hash tables. For an application, we achieved an image retrieval for kotenseki images. We wrote a comprehensive survey paper about PQ with Prof. Jegou who developed the original PQ algorithm.

Research Products

• [Int'l Joint Research] Facebook AI Research(France)
• [Journal Article] PQTable: Non-exhaustive Fast Search for Product-quantized Codes using Hash Tables (2018)
  • Author(s): Yusuke Matsui, Toshihiko Yamasaki, and Kiyoharu Aizawa
  • Journal Title: IEEE Transactions on Multimedia Volume: 99 Issue: 7 Pages: 1809-1822
  • DOI: 10.1109/tmm.2017.2774009
  • Peer Reviewed / Open Access
• [Journal Article] [Invited Paper] A Survey of Product Quantization (2018)
  • Author(s): Yusuke Matsui, Yusuke Uchida, Herve Jegou and Shin'ichi Satoh
  • Journal Title: ITE Transactions on Media Technology and Applications Volume: 6 Issue: 1 Pages: 2-10
  • DOI: 10.3169/mta.6.2
  • NAID: 130006301776
  • ISSN: 2186-7364
  • Peer Reviewed / Open Access / Int'l Joint Research
• [Journal Article] Sketch-based Manga Retrieval using Manga109 Dataset (2017)
  • Author(s): Yusuke Matsui, Kota Ito, Yuji Aramaki, Azuma Fujimoto, Toru Ogawa, Toshihiko Yamasaki, and Kiyoharu Aizawa
  • Journal Title: Multimedia Tools and Applications Volume: - Issue: 20 Pages: 21811-21838
  • DOI: 10.1007/s11042-016-4020-z
  • Peer Reviewed / Open Access / Int'l Joint Research / Acknowledgement Compliant
• [Presentation] Deep Image Retrieval Applied on Kotenseki Ancient Japanese Literature (2017)
  • Author(s): Chairath Sirirattanapol, Yusuke Matsui, Shin'ichi Satoh, Kuninori Matsuda, and Kazuaki Yamamoto
  • Organizer: IEEE International Workshop on Machine Learning and Computing for Visual Semantic Analysis (MLCSA)
  • Int'l Joint Research
• [Presentation] 古典籍に対する画像検索 (2017)
  • Author(s): 松井勇佑, 佐藤真一, 松田訓典, 山本和明
  • Organizer: 画像の認識・理解シンポジウム (MIRU)
• [Presentation] １時間で画像検索エンジンを作る (2017)
  • Author(s): 松井勇佑
  • Organizer: コンピュータグラフィックスとビジュアル情報学研究会
  • Place of Presentation: 東京
  • Invited
• [Remarks] Kotenseki Image Search
  • URL: http://yusukematsui.me/project/kotenseki/kotenseki.html
• [Remarks] A Survey of Product Quantization
  • URL: http://yusukematsui.me/project/survey_pq/survey_pq.html
• [Remarks] PQk-means
  • URL: http://yusukematsui.me/project/pqkmeans/pqkmeans.html
• [Remarks] Sketch-based Manga Retrieval
  • URL: http://yusukematsui.me/project/sketc2manga/sketch2manga.html
• [Remarks] Writing an Image Search Engine from Scratch
  • URL: http://yusukematsui.me/project/sis/sis.html