Prediction of pixels in an arbitrary region from pixels on the region boundary

被引:0
|
作者
Öktem, L [1 ]
Astola, J [1 ]
机构
[1] Tampere Univ Technol, Signal Proc Lab, FIN-33101 Tampere, Finland
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Solution of Dirichlet problem is proposed as a method for predicting the pixel values in an arbitrary region from the pixel values on the boundary of the region. Using the actual boundary pixel values as input to the predictor does not necessarily yield the best prediction. A computationally complex method is proposed for 'selecting' the pixel values on the boundary so as to yield an optimal prediction in a least squares sense. An iterative low complexity approximation to this method is also proposed. Experimental results show that both methods provide significant improvements overusing original boundary pixel values for prediction.
引用
收藏
页码:429 / 431
页数:3
相关论文
共 50 条
  • [41] Letter perception: from pixels to pandemonium
    Grainger, Jonathan
    Rey, Arnaud
    Dufau, Stephane
    TRENDS IN COGNITIVE SCIENCES, 2008, 12 (10) : 381 - 387
  • [42] From tiles to pixels: Media and the city
    Strickland, Roy
    PLACES-A FORUM OF ENVIRONMENTAL DESIGN, 2006, 18 (02): : 4 - 5
  • [43] Modelling cement microstructure: Pixels, particles, and property prediction
    D. P. Bentz
    Materials and Structures, 1999, 32 : 187 - 195
  • [44] Deep Hierarchical Planning from Pixels
    Hafner, Danijar
    Lee, Kuang-Huei
    Fischer, Ian
    Abbeel, Pieter
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [45] Unsupervised separation of dynamics from pixels
    Silvia Chiappa
    Ulrich Paquet
    METRON, 2019, 77 : 119 - 135
  • [46] Unsupervised separation of dynamics from pixels
    Chiappa, Silvia
    Paquet, Ulrich
    METRON-INTERNATIONAL JOURNAL OF STATISTICS, 2019, 77 (02): : 119 - 135
  • [47] From pixels to sentiment: Fine-tuning CNNs for visual sentiment prediction
    Campos, Victor
    Jou, Brendan
    Giro-i-Nieto, Xavier
    IMAGE AND VISION COMPUTING, 2017, 65 : 15 - 22
  • [48] From pixels to prognosis: Imaging biomarkers for discrimination and outcome prediction of pulmonary embolism
    Gotta, Jennifer
    Gruenewald, Leon D.
    Martin, Simon S.
    Booz, Christian
    Mahmoudi, Scherwin
    Eichler, Katrin
    Gruber-Rouh, Tatjana
    Biciusca, Teodora
    Reschke, Philipp
    Juergens, Lisa-Joy
    Onay, Melis
    Herrmann, Eva
    Scholtz, Jan-Erik
    Sommer, Christof M.
    Vogl, Thomas J.
    Koch, Vitali
    EMERGENCY RADIOLOGY, 2024, 31 (03) : 303 - 311
  • [49] Depth Boundary Reconstruction Method Using the Relations with Neighboring Pixels
    Kim, Donghyun
    Ryu, Seungchul
    Choi, Sunghwan
    Sohn, Kwanghoon
    2013 IEEE INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS (ICCE), 2013, : 466 - 467
  • [50] No-reference panoramic image quality assessment based on multi-region adjacent pixels correlation
    Huang, Xinpeng
    Liu, Xin
    Ding, Wenxin
    Meng, Chunli
    An, Ping
    PLOS ONE, 2022, 17 (03):