(3.238.235.155) 您好!臺灣時間:2021/05/11 19:12
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

: 
twitterline
研究生:吳佩真
研究生(外文):Wu, Pei-Chen
論文名稱:基於Kinect裝置之影像縮放技術
論文名稱(外文):Content-Based Image Retargeting using a Kinect Camera
指導教授:林維暘
指導教授(外文):Lin, Wei-Yang
口試委員:林維暘余松年黃春融
口試委員(外文):Lin, Wei-YangYu, Sung-NienHuang, Chun-Rong
口試日期:100/07/21
學位類別:碩士
校院名稱:國立中正大學
系所名稱:資訊工程研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2011
畢業學年度:99
語文別:中文
論文頁數:67
中文關鍵詞:影像大小調整深度資訊視覺顯著偵測高斯分佈不連續接縫切割動態執行
外文關鍵詞:Content-aware image resizingdepth informationsaliency detectionmean shiftgaussian distributiondiscontinuous seam carvingdynamic programming
相關次數:
  • 被引用被引用:2
  • 點閱點閱:1227
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:271
  • 收藏至我的研究室書目清單書目收藏:4
影像大小調整的技術在近年來廣泛地被應用,以往的做法只有比例上的縮放,但這種做法往往會導致主要物體的形變、失真。近年來提出的縮放技術是基於影像內容去做大小的調整,較為著名的方法為Avidan等人所提出的接縫方法(seam carving)。大部分這種對影像內容分析的技術僅限於對單一彩色(RGB)影像資訊做分析,而我們利用Kinect攝影機,同時擷取彩色資訊以及深度資訊,藉由分析這兩種影像,可以得到較佳的結果,能夠保護影像主體的部分,並且保持影像中的結構性,使得在影像縮放時將其失真程度降到最低。

本論文主要在展示深度資訊如何有效地改善接縫方法,與以往的做法不同,我們利用了深度資訊對梯度資訊、視覺顯著資訊等做非線性的組合,提出了基於深度的影像重要圖(depth-based importance map);其中,我們對於深度影像的處理方式是利用 mean shift 的找到 local peak ,此極值正好代表著重要物件的中心區域,並利用高斯分佈以此極值為中心,隨著距離而遞減物體的權重,也就是我們可以得到一個以重要物體為中心,往外遞減的權重分配圖。我們將這個基於深度的影像重要圖與接縫的能量圖以線性比例做結合,並且當作接縫方法的輸入。在接縫方法部分,參考了Grundmann等人提出的不連續接縫(discontinuous seam)的概念,將此概念加入本系統中,並以動態規劃方法(dynamic programming)執行。因此,藉由深度資訊及不連續接縫,本系統除了較能維持主要物體的完整性外,也更能保持影像的結構性,使得結果影像有較佳的視覺效果。
Imaging technology has made the capture and display of digital images ubiquitous. A variety of displays are used to view them, and it entails that resizing technique also be widely applied. Straightforward image resizing operators, such as scaling and cropping, often produce deformation and distortion. Recent retargeting approaches aim to resize images in a content-aware manner. The seam carving technique proposed by Avidan et al. is a popular approach in this field of study. Most retargeting approaches analyze only color information; we analyze both color and depth information captured by a Kinect to maintain the structure and preserve important regions.

In this thesis, we present the content-aware image retargeting algorithm with the help of depth information. We introduce a depth-based importance map, such that deformation of the image is guided by this map. This depth-based importance map is computed automatically using a novel combination of gradient, salience, and depth-based measures. Depth-based measures is a weight map, we analysis depth information and use mean shift procedure to find the local peak as the central of the important object, and apply Gaussian distribution to determine the weight. We apply this depth-based importance map as the
input of discontinuous seam carving algorithm and we compute the seam by dynamic programming method. To sum up, by adding the depth-based importance map in discontinuous seam carving algorithm, it visually behaves better in maintaining the structure and preserving important regions. In our experiments, the proposed method produces much better results than some existing approaches.
Contents

1 Introduction............................................1
1.1 研究背景與動機.........................................1
1.2 關於Kinect攝影機......................................4
1.3 論文架構..............................................6
2 Related Work............................................7
2.1 相關研究與應用.........................................7
3 Methods ............................................... 11
3.1 系統描述..............................................11
3.2 重要物件擷取...........................................13
3.2.1 影像擷取........................................13
3.2.2 深度影像處理.....................................14
3.2.3 視覺顯著區偵測...................................17
3.3 基於內容的影像大小調整...................................21
3.3.1 連續的接縫方法...................................21
3.3.2 不連續的接縫方法.................................23
4 Experiment Results......................................27
4.1 實驗一................................................28
4.2 實驗二................................................31
4.3 實驗三................................................34
4.4 實驗四................................................40
4.5 實驗五................................................43
4.6 實驗六................................................49
4.7 實驗七................................................52
4.8 實驗八................................................55
5 Conclusion...............................................59

[1] Goferman, S. and Zelnik-Manor, L. and Tal, A. Context-aware saliency detection. Computer Vision and Pattern Recognition (CVPR),pages 2376—2383, 2010.
[2] M. Rubinstein, A. Shamir, and S. Avidan, Improved seam carving for video retargeting, in ACM Transactions on Graphics. New York, USA: ACM, 2008, pp. 1--9.
[3] Grundmann, M., Kwatra, V., Han, M., Essa, I.: Discontinuous seam-carving for video retargeting. In: CVPR (2010)
[4] Wang, Y.-S., Tai, C.-L., Sorkine, O., and Lee, T.-Y., “Optimized scale-and-stretch for image resizing,” ACM Trans. on Graphics (Proc. of SIGGRAPH ASIA) 27(5) (2008).
[5] Y. Pritch and E. Kav-Venaki and S. Peleg, Shift-Map Image Editing, ICCV'09
[6] A. Mansfield, P. Gehler, L. Van Gool and C. Rother, Visibility Maps for Improving Seam Carving, ECCV 2010.
[7] 李仁豪,葉素玲. 選擇注意力:選工間或選物體? 應用心理研究,pages 165-194, 2004
[8] E. Fox. Perceptual grouping and visual selective attention. Attention, Perception and Psychophysics, 60(6):1004-1021,1998.
[9] M.R. Lamb and E.W. Yund. The role of spatial frequency in cued shifts of attention between global and local forms. Attention, Perception and Psychophysics, 62(4):753-761,2000.
[10] J.F. Juola, D.G. Bouwhuis, E.E. Cooper and C.B. Warner. Control of attention around the fovea. Journal of Experimental Psychology: Human Perception and Performance, 17(1):125-141,1991.
[11] D. Kahneman, A. Treisman and B.J. Gibbs. The reviewing of object files: Object-specific integration of information. Cognitive Psychology, 24(2):175-219,1992.
[12] J. Li and S. Yeh. Attentional capture without display-wide attentional setting: evidence from inattentional blindness. In European Conference on Visual Perception, 2002.
[13] B.A. Eriksen and C.W. Eriksen. Effects of noise letters upon the identification of a target letter in a nonsearch task. Attention, Perception and Psychophysics, 16(1):143-149,1974
[14] C.W. Eriksen and J.E. Hoffman. The extent of processing of noise elements during selective encoding from visual displays. Attention, Perception and Psychophysics, 14(1):155-160,1973.
[15] Games Blog. http://www.guardian.co.uk/technology/gamesblog/2010/jun/21/games-microsoft-kinect-pricing.
[16] PrimeSensor, http://www.primesense.tw/.
[17] Wiki, http://www.javaforge.com/wiki/103649.
[18] Avidan, S. and Shamir, A.. Seam carving for content-aware image resizing. ACM Transactions on Graphics, 26(3),2007.
[19] A. Mansfield, P. Gehler, L. Van Gool and C. Rother, Scene Carving: Scene Consistent Image Retargeting, ECCV 2010.
[20] C. H. Hsieh. Content-Based Video Retargeting with Depth Information. Master's thesis, Department of Electrical Engineering, National Chung Cheng University, Taiwan,2010.
[21] D. Vaqueroa, M. Turka, K. Pullib, M. Ticob, and N. Gelfandb. A survey of image retargeting techniques. Proceedings of SPIE the International Society for Optical Engineering, 779814(2010), 2010.
[22] P. Soille. Morphological image analysis: principles and applications. Springer-Verlag New York, Inc. Secaucus, NJ, USA, 2003.
[23] D. Comaniciu and P. Meer. Mean shift: A robust approach toward feature space analysis. IEEE Transactions on pattern analysis and machine intelligence, pages 603-619, 2002.
[24] Experimental Online Shift-Map Editing. http://www.cs.huji.ac.il/projects/shiftMapFlash/.


QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔