我們建立一個可以自由組合真實影像物體的系統，這個系統可以經由使用者隨意地擺設欲加入的物體影像，一部分是背景，並保留某些特徵(例：湖面的漣漪)或是含有前景的背景(例：空中的某?雲)，另一部分則是前景，並分類成三種，讓整個在加入前景的步驟，能夠讓使用者更彈性的組合影像，也讓在搜尋的過程中能找出更符合我們所輸入的圖片，好讓在鑲嵌的時候能更真實、更合乎道理。
接著藉由網路上的大量圖片當成資料庫，利用 imgSeek 和 mean shift clustering 的兩個步驟，在巨大的影像資料庫中找尋符合我們輸入影像的最佳圖像，最後在將前景圖片內嵌回到影像中之前，還加入了色彩空間轉移的步驟，讓從各個不同影像中取出的前後景，有著同樣的色系投影在其物體上，最後再參考背景圖光的來源，並幫新加入的物體加上相對應的影子，讓最後的影像能更加的真實。

We construct a system that freely composes many realistic part objects among different images. The system can be added any object from images by user. The objects in images are classified into two parts; one is ‘background’, keeps some features like the ripple on the lake or the background with meaningful foreground, another one is ‘foreground’, is assorted into three parts to make the process of adding foreground more flexible and make the result of searching in image database more realistic. Let the step in embedding image more reasonable.
Next step, we search the best matching image from the massive images on the Internet as the image database. And we provide two steps in searching filter to find out the best image. One is using the imgSeek software, another one is by mean shift clustering to filter out. Finally, before we embed the foreground object back to the best matching image, we transfer the color space into another image and add the relative shadow referenced to the light in candidate image to be the real image.

A. Agarwala, M. Dontcheva, M. Agarwala, S. Drucker, A. Colburn, B. Curless, D. Salesin, M. Cohen. Interactive Digital PhotoMontage. ACM Trans. Graph. 23, 3,
294-302. SIGGRAPH 2004.

W. A. Barrett , A. S. Cheney . Object-Based Image Editing. ACM Transactions on Graphics 21, 3, 777–784. 2002

M. Bertalmio , G. Sapiro , V. Caselles , C. Ballester. Image Inpainting. In Proceedings of ACM SIGGRAPH 2000,ACM Press / ACM SIGGRAPH, New-York, E. Fiume, Ed., ComputerGraphics Proceedings, Annual Conference Series, ACM, 417–424. 2000.

T. Chen , M. M. Cheng , P. Tan , A. Shamir , S. M. Hu PhotoSketch : Internet Image Montage. ACM Trans. Graphics (Proc. SIGGRAPH Asia 2009)

A. Criminisi , P. P ’erez , K. Toyama. Object Removal by Exemplar-based Inpainting. Proc. Conf. Computer Vision and Pattern Recognition, vol. 2, pp.
721-728. 2003.

N. Dialopoulos , I. Essa , R. Jain. Content based image synthesis. In Conference on Image and Video Retrieval. 2004.

I. Drori , D. Cohen-Or , H. Yeshurun. 2003. Fragment-Based Image Completion. SIGGRAPH 2003.

B. Geogescu , I. Shimshoni , P. Meer . Mean Shift Based Clustering in High Dimensions: A Texture Classification Example. In Proc. of ICCV, 456. 2003.

J. Hays , A. A.Efros. Scene completion using millions of photographs. SIGGRAPH 2007.

C. E. Jacobs , A. Finkelstein , D. H. Salesin. Fast Multiresolution Image Querying. SIGGRAPH 1995.

J. Jia , J. Sun , C. K. Tang , H. Y. Shum. Drag-and-Drop Pasting. Siggraph 2004.

Y. Li , J. Sun , C. K. Tang , H. Y. Shum. Lazy Snapping. SIGGRAPH 2004.

E. Mortensen , W . Barrett. Intelligent scissors for image composition. Proc. ACM Siggraph. 1995

P. P ’erez , M. Gangnet , A. Blake. Poisson image editing. SIGGRAPH 2003.

E. Reinhard , M. Ashikhmin , B. Gooch , P. Shirley. Color Transfer between Images. IEEE Computer Graphics and Applications, v.21 n.5, p.34-41, September 2001

C. Rother , V. Kolmogorov , A. Blake. “GrabCut” — Interactive Foreground Extraction using Iterated Graph Cuts. SIGGRAPH 2004.

A. Smeulders , M. Worring , S. Santini , A. Gupta. Content-Based Image Retrieval at the End of the Early Years. 2000.

J. Sun , L. Yuan , J. Jia , H. Y. Shum. Image Completion with Structure Propagation. SIGGRAPH 2005.

O. Tuzel, P. Meer. Mean Shift Clustering. 2002

J. Wang , M. Cohen. Simultaneous Matting and Compositing. In Proc. Of CVPR, 1-8. 2007.

ImgSeek , http://www.imgseek.net/ , a software for photo collection manager and viewer with content-based search and many other features.