Scene Carving: Scene Consistent Image Retargeting

被引:0
|
作者
Mansfield, Alex [1 ]
Gehler, Peter [1 ]
Van Gool, Luc [1 ,2 ]
Rother, Carsten [3 ]
机构
[1] ETH, Comp Vis Lab, Zurich, Switzerland
[2] Katholieke Univ Leuven, ESAT PSI, Leuven, Belgium
[3] Microsoft Res Ltd, Cambridge, England
来源
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
Image retargeting algorithms often create visually disturbing distortion. We introduce the property of scene consistency, which is held by images which contain no object distortion and have the correct object depth ordering. We present two new image retargeting algorithms that preserve scene consistency. These algorithms make use of a user-provided relative depth map, which can be created easily using a simple GrabCut-style interface. Our algorithms generalize seam carving. We decompose the image retargeting procedure into (a) removing image content with minimal distortion and (b) re-arrangement of known objects within the scene to maximize their visibility. Our algorithms optimize objectives (a) and. (b) jointly. However, they differ considerably in how they achieve tins. We discuss this in detail and present examples illustrating the rationale of preserving scene consistency in retargeting.
引用
收藏
页码:143 / +
页数:2
相关论文
共 50 条
  • [41] CROSS-SCENE HYPERSPECTRAL IMAGE CLASSIFICATION BASED ON CYCLE-CONSISTENT ADVERSARIAL NETWORKS
    Meng, Zhihao
    Ye, Minchao
    Yao, Futian
    Xiong, Fengchao
    Qian, Yuntao
    2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022), 2022, : 1912 - 1915
  • [42] PerspectiveNet: A Scene-consistent Image Generator for New View Synthesis in Real Indoor Environments
    Novotny, David
    Graham, Benjamin
    Reizenstein, Jeremy
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), 2019, 32
  • [43] Does consistent scene context facilitate object perception?
    Hollingworth, A
    Henderson, JM
    JOURNAL OF EXPERIMENTAL PSYCHOLOGY-GENERAL, 1998, 127 (04) : 398 - 415
  • [44] SceneScape: Text-Driven Consistent Scene Generation
    Fridman, Rafail
    Abecasis, Amit
    Kasten, Yoni
    Dekel, Tali
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [45] Improved seam carving for video retargeting
    Rubinstein, Michael
    Shamir, Ariel
    Avidan, Shai
    ACM TRANSACTIONS ON GRAPHICS, 2008, 27 (03):
  • [46] Stereo Seam Carving for Vertical Retargeting
    Gong Y.-Y.
    Guan B.-L.
    Zeng K.
    Li K.-H.
    Wu H.-F.
    Luo X.-N.
    Zeng, Kun (zengkun@gmail.com), 1600, Science Press (40): : 2589 - 2605
  • [47] FASTER SEAM CARVING FOR VIDEO RETARGETING
    Li, Chenyang
    Hu, Ruimin
    Liang, Chao
    Xiao, Chunxia
    Ruan, Weijian
    2018 25TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2018, : 823 - 827
  • [48] Creating Consistent Scene Graphs Using a Probabilistic Grammar
    Liu, Tianqiang
    Chaudhuri, Siddhartha
    Kim, Vladimir G.
    Huang, Qixing
    Mitra, Niloy J.
    Funkhouser, Thomas
    ACM TRANSACTIONS ON GRAPHICS, 2014, 33 (06):
  • [49] An Effective Method for Scene Image Management
    Wang Xiaochi
    Xu Jie
    Fang Zhigang
    2010 INTERNATIONAL CONFERENCE ON NETWORKING AND INFORMATION TECHNOLOGY (ICNIT 2010), 2010, : 217 - 220
  • [50] AN 'AFTER-IMAGE OF THE OPENING SCENE'
    NICOL, A
    NEW ENGLAND REVIEW AND BREAD LOAF QUARTERLY, 1986, 9 (02): : 183 - 183