SIFT

SIFT se yon sig pou Scale Invariant Feature Transform. Se yon metòd epitou yon kòd ki pèmèt swiv yon pwen, yon objè, yon fòm nan yon imaj, nan yon video. Li pèmèt gade reperaj yon bagay nan imaj an, menm si ge transfòmasyon sou li tankou translasyon, rotasyon, zoumen, chanjman echèl. Genyen anpil aplikasyon sou li.

Mèt li se David Lowe, yon chèchè ki mèt yon brevè si envansyon sa.

Yo rele lojisyèl sa yo detektè pou pwen enterèsman ( yon pwen enterèsman se yon pwen nan yon imaj ki prezante enfòmasyon aksesib nan imaj sa a)

Kòd

Chanjman echèl nan espas

Se nan operasyon sa ke SIFT ap detekte pwen enterèsman li. Li konvolye (wè konvolisyon) imaj an ak yon filt gosyen nan diferan echèl, li pran

Li itilize sa yo rele yon Diferans Gosyen (Gaussian)

Yon imaj ak Diferans Gosyen $D\left(x,y,\sigma \right)$ se

$D\left(x,y,\sigma \right)=L\left(x,y,k_{i}\sigma \right)-L\left(x,y,k_{j}\sigma \right)$,

nan kilès $L\left(x,y,k\sigma \right)$ se imaj nou an $I\left(x,y\right)$ konvolye avèk yon filt gosyen ki ap lise l (Gaussian blur) $G\left(x,y,k\sigma \right)$ at scale $k\sigma$, i.e.,

$L\left(x,y,k\sigma \right)=G\left(x,y,k\sigma \right)*I\left(x,y\right)$

Se pousa yon imaj avèk yon Diferans gosyen ant echèl $k_{i}\sigma$ ak $k_{j}\sigma$ se sèlman diferans ant imaj ak gosyen lise yo (Gaussian-blurred) nan echèl $k_{i}\sigma$ epi $k_{j}\sigma$.

Gade tou

• SURF
• GLOH
• chanjman echèl

Referans

• Lowe, D. G., “Distinctive Image Features from Scale-Invariant Keypoints”, International Journal of Computer Vision, 60, 2, pp. 91-110, 2004.
• Lowe, D. G., "Object recognition from local scale-invariant features," Proceedings of International Conference on Computer Vision, 1999, pp. 1150-1157
• Lindeberg, T., "Feature detection with automatic scale selection", International Journal of Computer Vision, 30:2, pages 77-116, 1998.
• Beis, J. and Lowe, D.G. 1997. "Shape indexing using approximate nearest-neighbor search in high dimensional spaces", Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Puerto Rico, pp. 1000-1006.
• Mikolajczyk, K. and Schmid, C. "A performance evaluation of local descriptors", IEEE Transactions on Pattern Analysis and Machine Intelligence, 10, 27, pp 1615--1630, 2005.
• Brown, M. and Lowe, D. G. "Recognising Panoramas", Proceedings of Ninth IEEE International Conference on Computer Vision, 2003. pp 1218- 1225 vol.2
• Se, S. Lowe, D. G. Little, J. "Vision-based Mobile Robot Localization And Mapping using Scale-Invariant Features", Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2001
• Iryna Gordon and David G. Lowe, "What and where: 3D object recognition with accurate pose," in Toward Category-Level Object Recognition, (Springer-Verlag, 2006), pp. 67-82
• Scovanner, P. Ali, S. Shah, M. "A 3-dimensional sift descriptor and its application to action recognition", Proceedings of the 15th International Conference on Multimedia, 2007. pp. 357-360
• Niebles, J. C. Wang, H. and Li, Fei-Fei. "Unsupervised Learning of Human Action Categories Using Spatial-Temporal Words", Proceedings of the British Machine Vision Conference (BMVC), Edinburgh, 2006

Lyen deyò

Kèk kòd ki itilize SIFT :

• Sebastian Nowozin (C#)
• Andrea Vedaldi (Matlab/C)
• Andrea Vedaldi (C++)
• David Lowe (C/Matlab)
• Rob Hess (C)
• Thomas El-Maraghi (Matlab)
• Dr Krystian Mikolajczyk (C)
• Stephan Saalfeld (Java)
• autopano-SIFT: a tool for generating panoramas from multiple images, using SIFT to match features between images