The features extracted for object recognition can be generally split
into two categories, local invariant and appearance-based. The
former is commonly selected for the recognition of generic objects,
while the latter is a popular choice for the recognition of specific
objects, for example, faces. Because most specific objects can be
aligned to appearance features, the deviations from the aligned
features offer some appearance characteristics good for recognition.
Such an alignment can be difficult to define in generic objects.
Therefore, the works on the recognition of generic objects using
appearance features in the literature are significantly outnumbered
by those using local invariant features. The performance of many
appearance features and associated classifiers on face recognition
has been widely studied and reported; however, their performance on
generic object recognition is only studied in a limited scope. To
extend our understanding in this regard and be able to determine the
appearance features and classifiers good for generic object
recognition, this paper reports a comprehensive comparison study in
which different combinations of features and classifiers are
evaluated on a benchmark database. To detect and segment the object
of interest from a scene, which is often the first step in object
recognition, we propose a scheme, called {\it Silhouette Alignment},
to align the features extracted from a test image to those in the
database. Although the appearance features considered in this study
are holistic, in the comparison we also include SIFT (Scale
Invariant Feature Transform), one of the most popular local
invariant features for object recognition, to justify the
performance of the appearance-based features and associated
classifiers. Experiments on the COIL-100 database show that DCT
features with Naive Bayesian classifier give the best performance
among others on object recognition across viewpoints. SIFT
outperforms most appearance features when the image quality is
degraded, i.e., blurred by noise. However, a few classifiers with
appearance features outperform SIFT in both noise-free conditions
and cluttered backgrounds.

The features extracted for object recognition can be generally split
into two categories, local invariant and appearance-based. The
former is commonly selected for the recognition of generic objects,
while the latter is a popular choice for the recognition of specific
objects, for example, faces. Because most specific objects can be
aligned to appearance features, the deviations from the aligned
features offer some appearance characteristics good for recognition.
Such an alignment can be difficult to define in generic objects.
Therefore, the works on the recognition of generic objects using
appearance features in the literature are significantly outnumbered
by those using local invariant features. The performance of many
appearance features and associated classifiers on face recognition
has been widely studied and reported; however, their performance on
generic object recognition is only studied in a limited scope. To
extend our understanding in this regard and be able to determine the
appearance features and classifiers good for generic object
recognition, this paper reports a comprehensive comparison study in
which different combinations of features and classifiers are
evaluated on a benchmark database. To detect and segment the object
of interest from a scene, which is often the first step in object
recognition, we propose a scheme, called {\it Silhouette Alignment},
to align the features extracted from a test image to those in the
database. Although the appearance features considered in this study
are holistic, in the comparison we also include SIFT (Scale
Invariant Feature Transform), one of the most popular local
invariant features for object recognition, to justify the
performance of the appearance-based features and associated
classifiers. Experiments on the COIL-100 database show that DCT
features with Naive Bayesian classifier give the best performance
among others on object recognition across viewpoints. SIFT
outperforms most appearance features when the image quality is
degraded, i.e., blurred by noise. However, a few classifiers with
appearance features outperform SIFT in both noise-free conditions
and cluttered backgrounds.

[1] K.Mikolajczyk and C.Schmid, “A performance evaluation of local descriptors,” TPAMI, vol. 27, no. 10, pp. 1615–1630, 2005.
[2] K.Mikolajczyk, T.Tuytelaars, C.Schmid, A.Zisserman, J.Matas, F.Schaffalitzky, T.Kadir, and L.J.V.Gool, “A comparison of affine region detectors,” IJCV, vol. 65, no. 1-2, pp. 43–72, 2005.
[3] M.Pontil and A.Verri, “Support vector machines for 3d object recognition,” TPAMI, vol. 20, no. 6, pp. 637 –646, jun 1998.
[4] A.M.Martinez and A.C.Kak, “Pca versus lda,” TPAMI, vol. 23, no. 2, pp. 228 –233, feb 2001.
[5] X.Liu, A.Srivastava, and K.Gallivan, “Optimal linear representations of images for object recognition,” TPAMI, vol. 26, no. 5, pp. 662 –666, may 2004.
[6] T.V.Pham and A.W.M.Smeulders, “Sparse representation for coarse and fine object recognition,” TPAMI, vol. 28, no. 4, pp. 555 –567, april 2006.
[7] A.Choksuriwong, B.Emile, H.Laurent, and C.Rosenberger, “Comparative study of global invariant descriptors for object recognition,” JEI, vol. 17, no. 2, pp. 023015, 2008.
[8] F.Rothganger, S.Lazebnik, C.Schmid, and J.Ponce, “3d object modeling and recognition using local affine-invariant image descriptors and multi-view spatial constraints,” IJCV, vol. 66, no. 3, pp. 231–259, 2006.
[9] X.Han, Y.Chen, and X.Ruan, “Multilinear supervised neighborhood embedding of local descriptor tensor for scene/object recognition,” TIP, vol. 21, no. 3, pp. 1314 –1326, 2012.
[10] C.Liu and H.Wechsler, “Gabor feature based classification using the enhanced fisher linear discriminant model for face recognition,” TIP, vol. 11, no. 4, pp. 467–476, April 2002.
[11] S.Lyu, “Mercer kernels for object recognition with local features,” in CVPR, 2005, vol. 2, pp. 223–229.
[12] X.H.Han, Y.W.Chen, and X.Ruan, “Multilinear supervised neighborhood em-bedding of a local descriptor tensor for scene/object recognition,” TIP, vol. 21, no. 3, pp. 1314 –1326, march 2012.
[13] C.Hua, Y.Mingqiang, K.Kidiyo, and R.Joseph, “Object recognition using relative-chord context,” in CECNet, april 2011, pp. 515 –518.
[14] M.Rudinac and P.P.Jonker, “A fast and robust descriptor for multiple-view object recognition,” in ICARCV, dec. 2010, pp. 2166 –2171.
[15] S A Nene, S K Nayar, and H Murase, “Columbia object image library (coil-100),” RLAAOTA, vol. 95, no. CUCS-006-96, pp. 223–303, 1996.
[16] I.Buciu, I.Nafornita, and I.Pitas, “Global gabor features for rotation invariant object classification,” in ICCP, 2008, pp. 41 –46.
[17] F.A.Pavel, Z.Wang, and D.D.Feng, “Reliable object recognition using sift fea¬tures,” in MMSP, oct. 2009, pp. 1 –6.
[18] P. Viola and M. Jones, “Rapid object detection using a boosted cascade of simple features,” in CVPR, 2001, vol. 1, pp. I–511 – I–518 vol.1.
[19] D.G.Lowe, “Distinctive image features from scale-invariant keypoints,” IJCV, vol. 60, pp. 91–110, 2004.
[20] D.Comaniciu, P.Meer, and Senior Member, “Mean shift: A robust approach toward feature space analysis,” TPAMI, vol. 24, pp. 603–619, 2002.
[21] M.J.Er, W.Chen, and S.Wu, “High-speed face recognition based on discrete cosine transform and rbf neural networks,” TNN, vol. 16, no. 3, pp. 679 –691, may 2005.
[22] J.-K.Kamarainen, V.Kyrki, and H.Kalviainen, “Invariance properties of gabor filter-based features-overview and applications,” TIP, vol. 15, no. 5, pp. 1088 –1099, 2006.
[23] C.Liu and H.Wechsler, “A gabor feature classifier for face recognition,” in ICCV, 2001, vol. 2, pp. 270 –275 vol.2.
[24] M.Lades, J.C.Vorbruggen, J.Buhmann, J.Lange, von der C.Malsburg, R.P.Wurtz, and W.Konen, “Distortion invariant object recognition in the dy¬namic link architecture,” TC, vol. 42, no. 3, pp. 300 –311, mar 1993.
[25] D.Price, S.Knerr, L.Personnaz, G.Dreyfus, and L.P.G.Dreyfus, “Pairwise neural network classifiers with probabilistic outputs,” in ANIPS 7. 1994, pp. 1109–1116, MIT Press.
[26] J.C. Platt, “Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods,” in ALMC. 1999, pp. 61–74, MIT Press.
[27] Kazuhiro Hotta, “Pose independent object classification from small number of training samples based on kernel principal component analysis of local parts,” IVC, vol. 27, no. 9, pp. 1240–1251, Aug. 2009.