Research
現在・過去の研究プロジェクトを紹介します
LINE-Drawn CGH
線状3Dオブジェクトのための高速ホログラフィ計算
点群モデルのCGH計算は点光源から発せられる球面波の重畳と解釈できます.同じ奥行に直線状に並んだ点群が作る波面が部分的に1次元的なパターンとして収束することを発見し,収束時の波面パターンで線状3DオブジェクトのCGHを高速作成する方法を考案しました.CPUのみでも十分な速度が出ることが特徴です.
The CGH calculation for a point cloud model can be interpreted as a superposition of spherical waves emitted from a point source. We discovered that wavefronts created by a point cloud arranged in a straight line at the same depth partially converge as a one-dimensional pattern, and devised a fast method to create CGH for linear 3D objects using the wavefront pattern at the time of convergence. This method is characterized by its ability to achieve sufficient speed even with only a CPU.
Reference
Takashi Nishitsuji, Tomoyoshi Shimobaba, Takashi Kakue, Tomoyoshi Ito, "Fast calculation of computer-generated hologram of line-drawn objects without FFT", Optics Express 28 (11), pp. 15904-15924 (2020. 5. 13).
David Blinder, Takashi Nishitsuji, Takashi Kakue, Tomoyoshi Shimobaba, Tomoyoshi Ito, Peter Schelkens, "Analytic computation of line-drawn objects in computer generated holography", Optics Express(IF:3.561), Vol.28, Issue.21, pp.31226-31240 (2020.10.02)
Takashi Nishitsuji, Takashi Kakue, David Blinder, Tomoyoshi Shimobaba, and Tomoyoshi Ito, "An interactive holographic projection system that uses a hand-drawn interface with a consumer CPU," Scientific Reports 11(1), 147 (2021.1.8).
Takashi Nishitsuji, David Blinder, Takashi Kakue, Tomoyoshi Shimobaba, Peter Schelkens, Tomoyoshi Ito., "GPU-accelerated calculation of computer-generated holograms for line-drawn objects", Optics Express Vol. 29, Issue 9, pp. 12849-12866 (2021.4.13).
David Blinder, Takashi Nishitsuji, and Peter Schelkens, "Real-time computation of 3D wireframes in computer-generated holography," IEEE Transactions on Image Processing (IF: 10.856), 30, 9418–9428 (2021.11.13).
Takashi Nishitsuji, Nobuya Shiina, David Blinder, Tomoyoshi Shimobaba, Takashi Kakue, Peter Schelkens, Tomoyoshi Ito and Takuya Asaka, "Variable-intensity line 3D images drawn using kinoform-type electroholography superimposed with phase error", Optics Express Vol. 30, Issue 15, pp. 27884-27902 (2022. 7. 15).
David Blinder, Takashi Nishitsuji and Peter Schelkens, "Three-dimensional spline-based computer-generated holography", Optics Express 31, 2, pp. 3072-3082 (2023.1.11).
Nobuya Shiina, Takashi Nishitsuji and Takuya Asaka, "Improving the imbalance of the light intensity of 3D wire-frame projection with electro-holography by superimposing a phase error", Optics Express 31 (23), pp. 37604-37617 (2023.10.24).
Takashi Nishitsuji, David Blinder, Tomoyoshi Shimobaba, Takashi Kakue, Peter Schelkens, and Tomoyoshi Ito, "Rapid calculation of computer-generated holograms for line-drawn 3D objects with varying thicknesses," Optics and Lasers in Engineering 181, 108359 (2024.6.13).
Press
"Lightning fast algorithms can lighten the load of 3D hologram generation", EurekAlert! (2020.6.6)
"Real "doodles of light" in real-time mark leap for holograms at home", EurekAlert! (2021.3.20)
"Hologram in daily life", The Statesman (Indian daily news paper) 14 Apr, p.11 (2021.4.15)
Funds
VQ-LUT
点群のベクトル量子化を用いた電子ホログラフィの圧縮伝送
奥行方向に連続する点群が作る波面が同心円状であることを活用し,点群を奥行方向にベクトル量子化することでCGH計算を高速化する手法を開発しました.量子化した点群に対応する波面の事前計算により,CGH計算の高速化と点群データ量の削減を両立に成功しました.電子ホログラフィの圧縮符号化方式として有望と考えています.
Utilizing the concentricity of wavefronts created by point clouds that are continuous in the depth direction, we have developed a method to speed up CGH calculations by vector quantizing the point clouds in the depth direction. By precomputing the wavefronts corresponding to the quantized point cloud, we succeeded in both speeding up the CGH calculation and reducing the amount of point cloud data. We believe that this method is promising as a compression encoding method for electron holography.
Reference
Takashi Nishitsuji, Yudai Hosono, Takashi Kakue, Tomoyoshi Shimobaba, Tomoyoshi Ito, Takuya Asaka, "Compression scheme of electro-holography based on the vector quantization of point light sources, " Optics Express Vol. 27, Issue 8, pp. 11594-11607 (2019. 4. 10).
Funds
HANDHELD PROJECTION
持ち運び可能な視覚的誘導デバイス
避難誘導時など多数の人々に対して情報を伝えようとするとき,音声は最もポピュラーな手段ですが,喧騒下では有効に働きません.他方,高速道路の速度低下対策として知られる光の流れ(ベクション)に見られるように,人間は直感的な視覚効果に誘導されることが知られています.本研究では,持ち運び可能な小型プロジェクタを用い,壁や天井などの構造物に動的な情報を投影することで,群衆等の効果的な誘導や情報提示の実現を目指しています.
When communicating information to many people, such as during an evacuation, voice is the most popular method, but it only works in a quiet environment. On the other hand, it is known that humans are guided by intuitive visual effects, such as light vection, which is known as a countermeasure for speed reduction on highways. This research aims to realize practical guidance of crowds and information presentation by using small, portable projectors to project dynamic information on walls, ceilings, and other structures.
References
佐久間 修平,西辻 崇,朝香 卓也,"携帯可能な小型プロジェクターを用いた投影型避難誘導システム", DICOMO2023, 2D-1 (2023.7.5)【ヤングリサーチャー賞】
佐久間修平,西辻 崇,朝香卓也,“屋内自己位置推定に基づく投影型ガイドシステム,” 2023年電子情報通信学会総合大会,D-9-5 (2023. 3. 7).
R2VS
全周囲視野のリアルタイム共有による映像監視システム
Realtime Remote Vision Sharing (R2VS)は,監視対象者の全周囲映像を遠隔地に伝送し,随伴するSPの如く,監視対象者の全周囲を高い臨場感の下で監視可能にするシステムです.固定型監視カメラの設置台数が増加する一方で,死角を完全に取り除くことは不可能です.そこで,個人の全周囲映像を共有することで,第三者の目をもって直感的に危険を察知し,対象者に伝える事ができます.ローカル5Gのような広帯域NWの下で実現したシステムです.
Realtime Remote Vision Sharing (R2VS) is a system that transmits all-surrounding images of a monitored subject to a remote location and enables monitoring of the subject's entire surroundings in a highly realistic manner, just like an accompanying SP. While the number of fixed surveillance cameras is increasing, it is impossible to eliminate blind spots. By sharing the all-surrounding image of an individual, it is possible to intuitively detect danger with the eyes of a third party and communicate it to the target. This system has been realized under a broadband NW such as local 5G.
References
乗松佑樹,加藤綾斗,西辻 崇,朝香卓也,“全天球カメラを用いた監視システムにおける監視映像品質と視聴環境の評価 ,” 電子情報通信学会コミュニケーションクオリティ研究会2023年5月研究会,CQ2023-6 (2023. 5. 19).
乗松佑樹,加藤綾斗,西辻 崇,朝香卓也,“ローカル 5Gを用いた高解像度全天球映像による遠隔監視システム,” 2023年電子情報通信学会総合大会,B-5-92 (2023. 3. 10).
加藤綾斗,西辻崇,朝香卓也,"ローカル5Gネットワークを用いた高解像度全天周映像伝送",映像情報メディア学会 メディア工学研究会,ME2022-65, オンライン (2022.7.22)
Ryoto Kato, Takashi Nishitsuji, Nobuya Shiina, Takuya Asaka, "Real-Time Field of View Sharing System based on Virtual Reality and 8K 360 Video with Private 5G Networks," 革新的無線通信技術に関する横断型研究会(MIKA) (2021.10.28).
Funds
東京都立大学ローカル5Gプロジェクト 挑戦型研究
Crape and Stamp method
ゾーンプレートの円対称性を利用したCGH高速計算
点群モデルのCGH計算は各点光源が発する球面波の重畳と解釈できます.球面波が作る波面は円対称性を持つことから,中心~外殻方向の軸上に分布する波面のみを定義通りに計算し,コンピュータグラフィクス技術における円描画手法により同心円状に展開することで,CGHの計算高速化とLUT法におけるメモリ量削減を実現しました.
The CGH calculation of a point cloud model can be interpreted as the superposition of spherical waves emitted by each point source. Since wavefronts created by spherical waves have circular symmetry, only wavefronts distributed along the axis from the center to the outer shell are calculated as defined, and then expanded into concentric circles using the circle drawing method in computer graphics technology.
Reference
Takashi Nishitsuji, Tomoyoshi Shimobaba, Takashi Kakue, Nobuyuki Masuda, Tomoyoshi Ito, "Fast calculation of computer-generated hologram using circular symmetry of zone plate," Optics Express , Vol. 20, Issue 25, pp. 27496-27502 (2012. 11. 27).
Takashi Nishitsuji, Tomoyoshi Shimobaba, Takashi Kakue, Tomoyoshi Ito, "Fast calculation of computer-generated hologram using run-length encoding based recurrence relation," Optics Express, Vol. 23, Issue 8, pp. 9852-9857 (2015. 4. 9).
TABLE less sin/cos for CGH
sin/cos関数のテーブルを用いない簡易近似手法
光伝搬計算が主のCGH計算において三角関数計算は頻繁に使用されます.ホログラフィ専用計算機HORNではsin/cos関数をテーブル実装していましたが回路密度のボトルネックになる問題がありました.そこで減算器とXORのみで構成可能&CGH向けには十分な精度で計算可能な三角関数の近似回路を開発,HORNの性能向上に貢献しました.
Trigonometric calculations are frequently used in CGH calculations, which are mainly light propagation calculations. HORN, a computer dedicated to holography, has a table implementation of sin/cos functions, but there was a problem of circuit density bottlenecks. We developed an approximate circuit for trigonometric functions that can be configured with only a subtractor and XOR, and that can be calculated with sufficient accuracy for CGH, thereby contributing to improved performance of HORN.
Reference
Takashi Nishitsuji, Tomoyoshi Shimobaba, Takashi Kakue, Daisuke Arai, Tomoyoshi Ito, "Simple and fast cosine approximation method for computer-generated hologram calculation," Optics Express Vol. 23, Issue 25, pp. 32465-32470 (2015. 12. 10).
Takashige Sugie, Takanori Akamatsu, Takashi Nishitsuji, Ryuji Hirayama, Nobuyuki Masuda, Hirotaka Nakayama, Yasuyuki Ichihashi, Atsushi Shiraki, Minoru Oikawa, Naoki Takada, Yutaka Endo, Takashi Kakue, Tomoyoshi Shimobaba, Tomoyoshi Ito, "High-performance parallel computing for next-generation holographic imaging," Nature Electronics Vol. 1, pp. 254-259 (2018. 4. 17).
Takashi Nishitsuji, Yota Yamamoto, Takashige Sugie, Takanori Akamatsu, Ryuji Hirayama, Hirotaka Nakayama, Takashi Kakue, Tomoyoshi Shimobaba, Tomoyoshi Ito, "Special-purpose computer HORN-8 for phase-type electro-holography," Optics Express Vol. 26, Issue 20, pp. 26722-26733 (2018. 9. 28).
Press
"A big step toward the practical application of 3D holography with high- performance computers", EurekAlert (2018.11.18)