Adapted SIFT Descriptor for Improved Near Duplicate Retrieval

Afra’a Ahmad Alyosef and Andreas N

urnberger

Department of Technical and Business Information Systems, Faculty of Computer Science,

Otto von Geruicke University Magdeburg, Magdeburg, Germany

Keywords:

Image Near Duplicate Retrieval, SIFT Descriptor, RC-SIFT 64D, Feature Extraction.

Abstract:

The scale invariant feature transformation algorithm (SIFT) has been designed to detect and characterize local

features in images. It is widely used to ﬁnd similar regions in afﬁne transformed images, to recognize similar

objects or to retrieve near-duplicates of images. Due to the computational complexity of SIFT based match-

ing operations several approaches have been proposed to speed up this process. However, most approaches

lack signiﬁcant decrease of matching accuracy compared to the original descriptor. We propose an approach

that is optimized for near-duplicate image retrieval tasks by a dimensionality reduction process that differs

from other methods by preserving the information around the keypoints of any region patches of the original

descriptor. The computation of the proposed Region Compressed (RC) SIFT−64D descriptors is therefore

faster and requires less memory for indexing. Most important, the obtained features show at the same time

a better retrieval performance and seem to be even more robust. In order to prove this, we provide results of

a comparative performance analysis using the original SIFT−128D, reduced SIFT versions, SURF−64D and

the proposed RC-SIFT−64D in image near-duplicate retrieval using large scale image benchmark databases.

1 INTRODUCTION

Finding similar images that show the same scene,

but have been taken with slightly different condi-

tions (so-called “near duplicate images”) is still a very

challenging task, even though its a very fundamental

problem in many real world tasks. In the literature

(e.g., (Xu et al., 2010), (Chum et al., 2007)) one can

meanwhile ﬁnd several benchmark databases contain-

ing sets of near duplicates that can be used to study

the performance of algorithms and similarity mea-

sures that have been designed for the task of ﬁnding

near duplicate images. The sets of near duplicate im-

ages provided slightly differ by quite diverse factors

as noise, blurring, compression rates, lighting condi-

tions or camera viewpoint. Therefore, these collec-

tions nicely cover characteristics of real world collec-

tions and tasks, e.g., such as the daily business of me-

dia agencies to ﬁlter and sort huge amounts of images

that have been taken by different photographers dur-

ing fashion fairs or political events. The main under-

lying problems here are data redundancy, e.g., having

two or more of copies of the same image in a collec-

tion; view conﬁguration issues, i.e., detecting near du-

plicate images which are taken using different conﬁg-

uration of a camera or different cameras; problems of

nipulated (fabricated) images and problems of time

offsets, i.e., ﬁnding similar images of the same scene

taken minutes or hours apart.

The ﬁrst step in image near-duplicate retrieval

(NDR) is to extract the features of an image. The

goal is to represent images by means of one or more

kinds of their distinct characteristics. One of the

most used approaches for ﬁnding the local features

in the ﬁeld of NDR (Auclair et al., 2006), (Chum

et al., 2008), (Zhang and et al., 2004), (Chu et al.,

2013) is the scale invariant feature transform algo-

rithm (SIFT) (Lowe, 2004), because the SIFT features

are invariant to scale and rotation variation and per-

form robustly even if the images differ in perspective,

noise, and illumination (Lowe, 2004). Furthermore,

approaches to index and quantize SIFT descriptors

have been proposed to improve the performance of the

matching process, especially in the image NDR tasks

(see e.g., (Jiang et al., 2015), (Nist

er and Stew

enius,

2006), (Auclair et al., 2006), (Chum et al., 2008)).

In this work, we propose a method to further

improve the performance of SIFT by reducing the

dimensionality of SIFT−128D descriptors to 64D.

Thereby we tackle two issues: First, especially in im-

age NDR ﬁeld the reduced SIFT−64D speeds up the

process of image matching by decreasing the mem-

ory and time complexity of the indexing and match-

Alyosef, A. and Nürnberger, A.

Adapted SIFT Descriptor for Improved Near Duplicate Retrieval.

DOI: 10.5220/0005694800550064

In Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods (ICPRAM 2016), pages 55-64

ISBN: 978-989-758-173-1

ing process. Second, the reduced SIFT improves

the robustness and accuracy of the matching pro-

cess. The performance of the proposed approach

is tested by conducting extensive experiments using

established benchmark datasets. The experiments

show that the reduced SIFT−64D improves the per-

formance of SIFT in image NDR tasks.

The remainder of this paper is organized as fol-

lows. Section 2 gives an overview of prior work re-

lated with the SIFT algorithm and image NDR algo-

rithms. Section 3 details the proposed method to re-

duce SIFT descriptors. Section 4 presents the settings

of our experiments and Section 5 discusses the results

of experiments. Finally, Section 6 draws conclusions

of this work and discusses possible future work.

2 RELATED WORK

Due to the robustness of the SIFT descriptor

against different kinds of image deformation, it

has been widely used in image NDR (Auclair

et al., 2006), (Chum et al., 2008), image classiﬁca-

tion (Nist

er and Stew

enius, 2006) and diagnosis of

tumors in medical images (Jiang et al., 2015).

In (Khan et al., 2011) the SIFT descriptor have

been reduced to 96D, 64D and 32D by ignoring the

contribution of some regions around the keypoints

when building the descriptors. It has been mentioned

in the original SIFT−128D algorithm that ignoring

some values of SIFT descriptor lead to decreased

matching performance. However, in (Khan et al.,

2011) SIFT−96D, 64D have shown robust perfor-

mance of image matching as the original SIFT−128D

across a substantial range of afﬁne transformation and

addition of noise. However, the performance of the

original SIFT−128D is still better than the perfor-

mance of SIFT−96D, 64D descriptors in case of ad-

ditional noise and illumination change (Khan et al.,

2011). In (Ke and Sukthankar, 2004) principle com-

ponent analysis is used to obtain the 64D SIFT de-

scriptors. In (J

egou et al., 2010) the SIFT descrip-

tors are aggregated to represent the images in form of

vectors after that, principle component analysis is ap-

plied to jointly optimize the dimensionality reduction

of these vectors. However, both approaches require a

training stage for the speciﬁc image collection.

To optimize the use of SIFT features in the match-

ing process, various techniques are suggested to struc-

ture, index and quantize the descriptors in suitable

form for further processing. In (Lowe, 2004) the

kd−tree has been used to structure the SIFT descrip-

tor in 128D space and to speed up their matching pro-

cess. However, the efﬁciency of kd−tree decreases

for high dimensional data because of the request

time for backtracking through the tree. In (Li et al.,

2014), (Yang and Newsam, 2008), (Grauman and

Darrell, 2005), (Grauman and Darrell, 2007), SIFT

descriptors have been quantized by splitting them into

k groups using the k−means clustering algorithm. In

this case a speciﬁc number of clusters is determined

and the descriptors are indexed by their closest cen-

ters. The produced cluster centers form a bag of

words and the images are presented in form of vectors

of these bag of words. In (Li et al., 2014), (Yang and

Newsam, 2008), (Grauman and Darrell, 2005), (Grau-

man and Darrell, 2007) the concept of bag of words is

used in further training steps to optimize the matching

process. In (Jiang et al., 2015), (Nist

er and Stew

enius,

2006) hierarchical k−means clustering is employed

to build a vocabulary tree. Each leaf node of the tree

is associated with an inverted ﬁle which contains the

indexes of images that have at least one descriptor

passed through a speciﬁc path in the tree. In our work,

the concept of a vocabulary tree is applied to index

the descriptors of several kinds. In the next subsec-

tion the details of compressing the SIFT descriptors

are explained.

3 REGION COMPRESSED SIFT

DESCRIPTOR FOR NDR

To motivate and describe our suggested modiﬁcations

of the SIFT descriptor, we will ﬁrstly explain brieﬂy

the working mechanism of the SIFT detector and de-

scriptor (Lowe, 2004).

3.1 SIFT−128D Descriptor

The detection of SIFT features can be achieved in four

main stages speciﬁed as follow: scale space extrema

detection, keypoint localization, orientation computa-

tion and keypoint descriptor computation.

In the ﬁrst stage image scale space is built by

downsampling and blurring the input image several

times. The blurring is achieved by convolving the in-

put image with multiple-scale Gaussian ﬁlters. After

that, the difference of neighbors in the scale space is

computed to form difference of Gaussian (DoG) im-

ages. In the second stage, SIFT keypoints are deter-

mined by ﬁnding the local maxima and minima in

DoG images. The stability of keypoints is veriﬁed

against contrast change and edge response and the un-

stable keypoints are rejected. In the third stage, the

dominant orientation is determined and assigned to

each keypoint.

ICPRAM 2016 - International Conference on Pattern Recognition Applications and Methods

In the ﬁnal stage a highly distinctive descriptor is

computed at each keypoint. The SIFT descriptor is

extracted from the region around the keypoint which

is called region of interest (RoI). The RoI is rotated

around a keypiont relative to the dominant orienta-

tion. Afterwards, a n ×n orientation histogram is cre-

ated over the RoI. For each bin in the histogram r ori-

entations are assigned, so that the descriptor has three

dimensions and n × n × r element. The size of the

SIFT descriptor is controlled by the width of the ori-

entation histogram n and the number of orientation

bins r. In the original SIFT algorithm (Lowe, 2004),

it has been shown that the best matching results are

reported when n = 4 and r = 8, i.e., when a descriptor

of 4 × 4 × 8 = 128 element is constructed. Fig. 1(a)

presents the way in which the SIFT−128D descriptor

is constructed.

However, the high dimensionality of SIFT de-

scriptor (128D) increases the sparsity of descriptors

and this may affect the accuracy of descriptor index-

ing in image NDR (for a discussion of problems re-

lated to high dimensional data indexing and cluster-

ing, see e.g., (Steinbach et al., 2003)). Therefore, in

this work we aim to compress the dimensionality of

the SIFT descriptor. In the next subsection, we ex-

plain our approach to compress SIFT descriptor.

3.2 Region Compressed SIFT

Descriptor

To increase the efﬁciency of descriptor indexing, i.e.,

speeding up the process and reducing the amount of

stored data in NDR, we propose a method to com-

press the dimensionality of the SIFT descriptor from

128D to 64D. We achieve this by extracting ﬁrst the

SIFT features in the same way as in the original SIFT

algorithm (Lowe, 2004) (as described in Section 3.1).

Afterwards, the descriptors are computed over all pix-

els in RoI with speciﬁc location, gradient and orienta-

tion with respect to the corresponding keypoint. The

descriptor is computed in form of a 3D histogram cen-

tered at the keypoint. In the original SIFT algorithm,

this descriptor has the dimensions 4 × 4 × 8. The val-

ues of these three dimensions indicate how the key-

point can be shifted to each allowed position in RoI

in vertical and horizontal locations that is 4 × 4 lo-

cations. For each location 8 directions are allowed

between 0

and 360

. In contrast to the reduction

method presented in (Khan et al., 2011) through ig-

noring some patches of RoIs, we suggest in this work

that for each two possible horizontal shifting in the

same direction with respect to the keypoint, only one

vertical shifting is available so that, for the all possi-

ble horizontal shifting (i.e., four horizontal shifting)

in all directions only two vertical shifting exists. For

each of this (4 × 2) locations eight directions are as-

signed. In this way we reduce the amount of the pos-

sible change of the SIFT descriptor when the RoI is

modiﬁed. Moreover, the number of altered bins in

the RoI histogram decreases. As a result we obtain

4×2 ×8 histogram i.e., 64D SIFT descriptor. We call

our method for extracting and compressing SIFT de-

scriptor ”Region Compressed SIFT” (RC-SIFT). The

histogram at each keypoint can be presented by a

triplet of elements H

, H

and H

where:

= y −

− 1

(1)

= x −

− 1

(2)

2π

(3)

where N

and N

deﬁne the number of bins in H

and

, respectively. The values of y and x are deﬁned

as y = 0, ..., N

− 1, and x = 0, ..., N

− 1. N

is the

number of orientations in each bin of the histogram

and θ is deﬁned as θ = 0, ..., N

− 1.

In this work, we perform the experiments for N

2, N

= 4 and N

= 8 to get the descriptor of the

form 4 × 2 × 8. We refer to this descriptor as RC-

SIFT−64D(R) (see Figure 1(b)). Afterwards, the ex-

periment is applied for N

= 4, N

= 2 and N

= 8

to obtain the descriptor of the form 2 × 4 × 8. we re-

fer to this descriptor as RC-SIFT−64D(C) (see Fig-

ure 1(c)). After that, experiments are performed for

= 2, N

= 2 and N

= 8 to get RC-SIFT−32D

and ﬁnally for N

= 2, N

= 2 and N

= 4 to get RC-

SIFT−16D.

In this way the compressed SIFT descriptor pre-

serves the size of RoI around a keypoint, i.e., con-

trary to the method suggested in (Khan et al., 2011),

no region around the keypoint is ignored. In the next

step the efﬁciency of RC-SIFT−64D, RC-SIFT−32D

and RC-SIFT−16D are evaluated against the perfor-

mance of the original SIFT−128D, SURF−64D and

SIFT−64D suggested in (Khan et al., 2011) for image

near-duplicate retrieval.

3.3 SIFT Descriptors Indexing with a

Vocabulary Tree

The straightforward way to match SIFT features is

exhaustive search, which can be achieved by match-

ing each feature of a given query image with all fea-

tures in the feature database. However, the exhaus-

tive search of SIFT features is extremely time con-

suming especially for large scale image databases

Adapted SIFT Descriptor for Improved Near Duplicate Retrieval

(a) 4 × 4 computed orientation histogram

array in SIFT−128D.

(b) 2 × 4 computed orientation his-

togram array in RC-SIFT−64D(R).

array in RC-SIFT−64D(C).

Figure 1: Comparison between SIFT−128D and RC-SIFT−64D descriptors. We refer to the compression of forms 4 × 2 × 8

and 2 × 4 × 8 as RC-SIFT−64(R)and RC-SIFT−64(C), respectively. These symbols are used in the all presented tables.

which produce a huge amount of features. To over-

come this problem, hashing functions (Auclair et al.,

2006), (Chum et al., 2008), direct clustering (Chum

et al., 2007), (Zhang et al., 2013) and hierarchical

clustering (Jiang et al., 2015), (Nist

er and Stew

enius,

2006) have been adapted to quantize and index SIFT

descriptors. In our study, a vocabulary tree and in-

verted ﬁles are used as described in (Jiang et al.,

2015), (Nist

er and Stew

enius, 2006) to index SIFT de-

scriptors. The vocabulary tree is built by applying the

k−means algorithm on the entire descriptor database

which split them into k clusters where each cluster

consists of a set of descriptors closest to a particular

center. This process is applied recursively on each

cluster to build a vocabulary tree of depth L and k

leaf nodes. The tree nodes present cluster centers and

are referred to as ”visual word” (Jiang et al., 2015).

The leaf nodes in the tree are represented by inverted

ﬁles. Each inverted ﬁle contains the indexes of the

images that they represent with at least one descrip-

tor at a particular leaf node. The inverted ﬁles of the

leaf nodes are concatenated to get the inverted ﬁles of

inner and root nodes. These inverted ﬁles strongly

speed up the matching process. Moreover, the in-

verted ﬁles help to adapt weights for the branches of

the tree.

In (Nist

er and Stew

enius, 2006) the L1−norm and

L2−norm have been used to compute the similarity

between images. The L1−norm tends to give better

matching results (Nist

er and Stew

enius, 2006). In our

evaluation, we use both the L1−norm and L2−norm

in order to compute the similarity between normalized

query and database vectors by traversing each vector

in a vocabulary tree as described in Eq. 4 (Nist

er and

Stew

enius, 2006).

3.4 Complexity of Vocabulary Tree for

SIFT−64D and SIFT−128D

After building the vocabulary tree (see Section 3.3),

the tree is used for image matching. The complexity

of this process is computed assuming that descriptors

are represented as character data-type. A D dimen-

sional descriptor vocabulary tree of depth L and k

leaf nodes need a memory of O(Dk

). Speciﬁcally,

a 128D descriptor tree requires O(128k

) whereas,

a 64D descriptor tree requires only O(64k

). More-

over, the time complexity of building a vocabulary

tree is affected by the dimensionality of descriptors.

Considering that total number of nodes in the vo-

cabulary tree is given as Σ

i=1

L+1

−k

k−1

≈ k

(see

also (Nist

er and Stew

enius, 2006)), the time complex-

ity of the vocabulary tree for a D− dimensional de-

scriptor database is given as O(DNT k

), where k is

the number of initial clusters, T is the iteration of al-

gorithm and N is the number of all descriptors of a

given image database. Based on this, the time com-

plexity of building a tree for a descriptor database

of dimensionality D = 128 is O(128NT k

) whereas

the time complexity for the RC-SIFT−64D descrip-

tors is O(64NT k

). So it is just a linear decrease

but the suggested RC-SIFT−64D descriptors obvi-

ously speeds up the indexing process and reduce the

required memory for processing. Table 1 presents

that the indexing time needed by RC-SIFT−64D and

SIFT−64D (Khan et al., 2011) is about the halve time

needed by SIFT−128D. The presented results are

computed using a vocabulary tree of depth L = 4 and

initial centers k = 10.

In the next section, the performance of the sug-

gested RC-SIFT (64D, 32D and 16D) descriptors

are evaluated against the performance of the original

SIFT−128D, SIFT−64D and SURF−64D based on

typical near-duplicate retrieval tasks.

ICPRAM 2016 - International Conference on Pattern Recognition Applications and Methods

Table 1: The computation time needed to perform the in-

dexing for both SIFT−128D, RC-SIFT−64D and SIFT-

64D (Khan et al., 2011) using a standard processor(Intel(R)

Core(TM)i5-2500 CPU) and a Matlab implementation.

Desc-No refers to the number of descriptor vectors.

Method Time(sec) Desc-No

SIFT-128D 3396.86 2, 095, 545

RC-SIFT-64D 1639.82 2, 095, 545

SIFT-64D 1588.38 2, 095, 545

4 EVALUATION

In the image near-duplicate retrieval ﬁeld, the per-

formance of the RC-SIFT−64D, RC-SIFT−32D and

RC-SIFT−16D is veriﬁed against the SIFT−128D,

the SURF−64D descriptor and the SIFT−64D (Khan

et al., 2011) descriptor mentioned in 2. The perfor-

mance is measured on a large scale image databases

using the vocabulary tree for feature indexing and

L1−norm to achieve the image NDR task. The vo-

cabulary trees are constructed as described in Subsec-

tion( 3.3) for each kind of the used descriptors sepa-

rately. In our experiment the initial number of clusters

is k = 10.

To perform the NDR task, the similarity between

normalized query vectors q img and database vectors

d img is computed by traversing each vector in a vo-

cabulary tree and it is given as (Nist

er and Stew

enius,

2006):

s(q img, d img) =



q img

−

d img



(4)

The normalization can be in any desired norm. In our

experiment L1−norm and L2−norm.

In this work we used our own implementation

of SIFT algorithm using some ”Opencv” functions.

SURF descriptors are computed by means of Opencv

functions. The vocabulary tree is constructed using

Matlab functions and VLFeat library. Moreover, we

implement the SIFT−64D described in (Khan et al.,

2011) based on our implementation of SIFT algorithm

by ignoring some patches of descriptors as it is de-

scribed in (Khan et al., 2011).

Furthermore, the experiment is performed on

benchmark databases as used in (Nist

er and

Stew

enius, 2006) (the dataset can be download

from the website (Nist

er and Stew

enius, )). These

databases consist more than 10, 000 of indoor/outdoor

images for about 2, 500 different scenes. The images

of each scene differ through a combination of changes

(additional blurring, scale change, rotation change, il-

lumination decrease/increase, additional noise, view-

point change) and another conditions (i.e., appear new

objects and occlusion of objects).

The results of the experiments are evaluated by

computing the recall value. Since we always have a

ﬁxed number of relevant images and the comparison

is done use a ranked list of ﬁxed length (i.e., length of

one, three or ten images as indicated in the respective

tables), we omit precision since it is directly corre-

lated to the recall values. Considering N

is the num-

ber of relevant images to a speciﬁc query image in the

database, N

the number of relevant images obtained

in matching results, then the recall is deﬁned as fol-

lows:

Recall =

(5)

5 RESULT AND ANALYSIS

The results of SIFT−64D and SIFT−128D are eval-

uated in different cases using various kinds of image

databases.

5.1 Mixed Database

The database described in (Nist

er and Stew

enius,

2006) is used for experiments. This database con-

tains four different images of 2, 550 different in-

door/outdoor scenes i.e., 10, 200 images in total.

Moreover, it contains a mixed complex images for

some scenes (i.e., image of the same scene but present

different arrangement of objects, appear/disappear of

some objects in addition to changes in lightness, con-

trast, sharpness, scale, and viewpoint conditions). To

test the robustness of the RC-SIFT in image NDR

ﬁeld, we select the ﬁrst image of each scene as a query

image while the remaining three images of each scene

are used as a basic database for retrieval task (i.e.,

2, 550 query image and 7, 550 database image). The

features and descriptors are extracted using the orig-

inal SIFT−128D, SURF−64D, SIFT−64D (Khan

et al., 2011) (produced by ignoring some patches of

the RoIs) and our RC-SIFT−64D, RC-SIFT−32D

and RC-SIFT−16D. After that, the descriptors of

each kind are indexed separately using the vocabu-

lary tree of depth L = 4 and initial clusters k = 10.

To achieve the retrieval task, L1−norm and L2−norm

are used to compute the distance between the query

image and the database images as described in equa-

tion 4. However, in our experiment L1−norm ob-

tains better results than L2−norm. A query image

is considered to be retrieved if its corresponding im-

ages in the database appear in the top three or ten

retrieved images. Table 2 summarizes the results

Adapted SIFT Descriptor for Improved Near Duplicate Retrieval

of the all proposed kinds of descriptors. It shows

that the RC-SIFT−64D obtained slightly better re-

sults than SIFT−128D. However, it presents that the

RC-SIFT−64D obtains better results than the RC-

SIFT−32D and RC-SIFT−16D. We assume that the

ignoring of some patches of the original descriptors

affects the performance of SIFT in solving image

NDR task therefore, the performance of RC-SIFT64D

seems to be superior than the performance of the sug-

gested SIFT−64D (Khan et al., 2011). The perfor-

mance of SURF descriptor seems to be low in solving

image NDR task. In our evaluation, it seems to be re-

lated with the complexity of the scenes. This result is

consistent with other studies, e.g., (Khan et al., 2011).

However, in (Khan et al., 2011) only a subset of

a benchmark database (Nist

er and Stew

enius, 2006)

(i.e., 2, 500 images) are used whereas, our experiment

is applied on the whole images in the same database

(i.e., 10, 200 images). The worst results are obtained

when the dimensionality of RC-SIFT is compressed

to 16D. Figure 2 presents a comparison between the

all proposed methods to achieve image NDR task. It

shows that the best results are found when the RC-

SIFT−64D is used. However, there is another exam-

ples where the SIFT−128D preforms the best. More-

over, we note in many cases that despite the equiva-

lent results of SIFT−128D and RC-SIFT−64D obtain

better ranking of the results than SIFT−128D. Fig-

ure 3 presents an example where the performance of

SIFT−128D and RC-SIFT−64D is equivalent but the

ranking of the results found by RC-SIFT−64D is bet-

ter than SIFT−128D.

The experiment is repeated for vocabulary trees of

depth L = 1, 2, 3, 4 and the best retrieval results for all

proposed descriptors are obtained when L = 4. There-

fore, we present the results only when a vocabulary

tree of depth L = 4 is used.

In the next step the robustness and invariant prop-

erties of the RC-SIFT−64 are veriﬁed against dif-

ferent image transformations, blurring, scale change

and viewpoint change. These invariant and robust-

ness properties are compared for the all proposed de-

scriptors. We don’t verify the invariant and robust-

ness properties of RC-SIFT−32 and RC-SIFT−16

because their performance in solving NDR task is low

comparing with the other used descriptors.

5.2 Image Afﬁne Transformations

To verify the robustness of our RC-SIFT against dif-

ferent kind of image transformations in the ﬁeld of

image NDR, the performance of all proposed descrip-

tors is evaluated against rotation change, illumina-

tion increase or decrease and adding different kinds of

Table 2: The retrieval performance of SIFT−128D,

SIFT−64D, SURF−64D and our RC-SIFT-64D, RC-SIFT-

32D and RC-SIFT-16D using a large ground truth database

(7, 650 images) with groups of three images belong to the

same scene and using a set of 2, 550 query images, each of

them has three related images in the database. The recall

is computed ﬁrstly based on the top three retrieved images

and secondly using the top ten retrieved images. The sym-

bols RC-SIFT−64D(R) and RC-SIFT−64D(C) are used to

refer for the compression of forms 4 × 2 × 8 and 2 × 4 × 8,

respectively.

Method Top 3 results Top 10 results

SIFT-128D 0.4932 0.5797

SIFT-64D 0.2715 0.3534

SURF-64D 0.2432 0.2952

RC-SIFT64D(R) 0.5067 0.6013

RC-SIFT64D(C) 0.4989 0.5914

RC-SIFT-32D 0.2892 0.3365

RC-SIFT-16D 0.2460 0.2878

noise. To achieve this, 500 images of different scene

of the benchmark database (Nist

er and Stew

enius,

2006) are picked to test the invariance properties of

features. The setting of generating the transformed

images are similar to the setting applied in (Khan

et al., 2011). The descriptors are indexed using a vo-

cabulary tree of depth L = 4 and initial centers k = 10.

The similarity is computed using L1−norm. A query

image is considered to be found in the database if its

corresponding database image appear in the top of the

retrieved images.

5.2.1 Rotation Change

To verify the rotation invariance, the ﬁrst 500 images

of the benchmark database (Nist

er and Stew

enius,

2006) are rotated at different angels in a clockwise

direction to generate 500 database images for each

angle. Results of NDR task are summarized in Ta-

ble 3. The results shows that all proposed descriptors

(SIFT−128D, SIFT−64D, SURF−64D and our RC-

SIFT-64D) are rotation invariant. For a big rotation

change the results shows that RC-SIFT-64D perform

a little bit better than the other proposed descriptors.

5.2.2 Addition of Noise

To test noise invariance, three types of noise are ap-

plied for the ﬁrst 500 images of the database (Nist

and Stew

enius, 2006). These types are: Gaussian

noise, salt and pepper noise and multiplicative noise.

The noise is added to the images using the follow-

ing settings: Gaussian white noise with σ

= 0.1 and

ICPRAM 2016 - International Conference on Pattern Recognition Applications and Methods

(a) Query image

(b) The top three results found by SIFT−128D

(d) The top three results found by SURF−64D

(e) The top three results found by RC-SIFT−64D

Figure 2: An example of the results of all proposed meth-

ods to perform the image NDR task. In this example RC-

SIFT−64D shows the best matching results.

Table 3: The performance comparison of SIFT−128D,

SIFT−64D, SURF−64D and our RC-SIFT-64D(R)and RC-

SIFT-64D(C) using a ground truth database (500 images)

and a set of 500 query images, each of them has one rotated

image in the database. For each query image we check if its

corresponding database image appear as the ﬁrst retrieved

image in the result. The experiment is repeated for the rota-

tion values:

{

, 135

, 215

, 250

}

Method 40

135

215

250

SIFT-128D 0.934 0.926 0.934 0.918

SIFT-64D 0.928 0.924 0.926 0.928

SURF-64D 0.933 0.926 0.933 0.920

RC-SIFT64 (R) 0.931 0.924 0.930 0.924

RC-SIFT64 (C) 0.928 0.924 0.928 0.923

= 0.2, salt and pepper noise with density of 15%

and 35% and multiplicative white noise with mean

0 and σ

= 0.04. The performance of the all used

(a) Query image

(b) The top three query results by SIFT−128D

Figure 3: Comparison of retrieval results of SIFT−128D

and RC-SIFT−64D descriptors. This example shows

an equivalent performance of the RC-SIFT−64D and

SIFT−128D in retrieving the belonged images to the same

scene. However, RC-SIFT−64D presents better raking of

the results than SIFT−128D.

descriptor in this work are presented in Tables 4, 5

and 6. These results show that performance all pro-

posed descriptors decrease very strongly when the ra-

tio of noise incease (see Tables 4 and 5). However, in

case of using the salt and pepper noise RC-SIFT-64D

obtains better results than the other descriptors even

though when the ratio of noise increases.

Table 4: The performance of SIFT−128D, SIFT−64D,

SURF−64D and our RC-SIFT−64D using a ground truth

database (500 images) and a set of 500 query images, each

of them has one Gaussian noised image in the database. The

experiment is repeated for σ

= 0.1 and σ

= 0.2. For each

query image we check if its corresponding database image

appear as the ﬁrst retrieved image in the result.

Method σ

= 0.1 σ

= 0.2

SIFT-128D 0.684 0.356

SIFT-64D 0.648 0.290

SURF-64D 0.644 0.352

RC-SIFT-64D(R) 0.682 0.352

RC-SIFT-64D(C) 0.679 0.354

5.2.3 Illumination Change

The illumination invariance is veriﬁed in the cases of

increase and decrease the brightness of the 500 test

Adapted SIFT Descriptor for Improved Near Duplicate Retrieval

Table 5: The performance of SIFT−128D, SIFT−64D,

SURF−64D and our RC-SIFT-64D using a ground truth

database (500 images) and a set of 500 query images,

each of them has one salt and pepper noised image in the

database. For each query image we check if its correspond-

ing database image appear as the ﬁrst retrieved image in the

result. The experiment is performed for two level of noise

density (i.e., 15% and 35%).

Method 15% 35%

SIFT-128D 0.826 0.202

SIFT-64D 0.822 0.152

SURF-64D 0.812 0.145

RC-SIFT-64D(R) 0.834 0.208

RC-SIFT-64D(C) 0.831 0.205

Table 6: The performance of SIFT−128D, SIFT−64D,

SURF−64D and our RC-SIFT-64D using a ground truth

database (500 images) and a set of 500 query images, each

of them has one multiplicative noise noised image in the

database. For each query image we check if its correspond-

ing database image appear as the ﬁrst retrieved image in the

result.

Method σ

Recall

SIFT-128D 0.04 0.98

SIFT-64D 0.04 0.822

SURF-64D 0.04 0.801

RC-SIFT-64D(R) 0.04 0.972

RC-SIFT-64D(C) 0.04 0.970

images. This is done by adding or subtracting a value

of all pixel’s channels(i.e., the channels red, green and

blue of each pixel are incremented equally). The val-

ues of pixel’s channels are adjusted to be within the

range 0 − 255. The brightness effect is tested using

the values

{

50, 70, 100, 120

}

and the darkness effect

is test using the values

{

−30, −50, −70, −90

}

. Re-

sults of NDR tasks summarized in Tables 7 and 8 de-

scribe that all kinds of used descriptors perform well

for illumination increase and decrease.

5.3 Image Blurring

To test the robustness of descriptors against image

blurring, three blurred image databases are gener-

ated using the ﬁrst 500 images of the benchmark

database (Nist

er and Stew

enius, 2006) and using three

different values of Gaussian ﬁlter i.e., σ

= 5, σ

= 10

and σ

= 20. Table 9 shows that the performance de-

grade most clearly when the ratio of blurring increase

(i.e., when the value of σ

increases). However, for

a small amount of blurring the descriptors seem to be

invariant. For a big amount of blurring, RC-SIFT-64D

is superior in matching images.

Table 7: The retrieval performance of SIFT−128D,

SIFT−64D, SURF−64D and our RC-SIFT-64D using a

ground truth database (500 images) and a set of 500 query

images, each of them has one brightened image in the

database. For each query image we check if its correspond-

ing database image appear as the ﬁrst retrieved image in the

result. The results are checked for the following brightness

values:

{

50, 70, 100, 120

}

Method 50 70 100 120

SIFT-128D 1.00 1.00 0.968 0.912

SIFT-64D 0.998 0.998 0.942 0.902

SURF-64D 0.995 0.993 0.932 0.881

RC-SIFT64(R) 0.995 0.995 0.956 0.902

RC-SIFT64(C) 0.970 0.970 0.956 0.902

Table 8: The retrieval performance of SIFT−128D,

SIFT−64D, SURF−64D and our RC-SIFT-64D using a

ground truth database (500 images) and a set of 500 query

images, each of them has one darkened image in the

database. The performance is presented using the darkness

values:

{

−30, −50, −70, −90

}

. For each query image we

check if its corresponding database image appear as the ﬁrst

retrieved image in the result.

Method −30 −50 −70 −90

SIFT-128D 1.00 0.992 0.958 0.810

SIFT-64D 0.998 0.992 0.956 0.802

SURF-64D 0.997 0.992 0.947 0.822

RC-SIFT64(R) 1.00 0.990 0.955 0.822

RC-SIFT64(C) 0.997 0.988 0.955 0.820

Table 9: Comparison of retrieval performance of

SIFT−128D, SIFT−64D, SURF−64D and our RC-SIFT-

64D using a ground truth database (500 images) and a set

of 500 query images, each of them has one blurred image

in the database. For each query image we check if its corre-

sponding database image appear as the ﬁrst retrieved image

in the result. The experiment is repeated for different level

of blurring using σ = 5, σ = 10 and σ = 20.

Method σ = 5 σ = 10 σ = 20

SIFT-128D 0.922 0.426 0.358

SIFT-64D 0.836 0.368 0.330

SURF-64D 0.859 0.347 0.298

RC-SIFT-64D(R) 0.830 0.366 0.386

RC-SIFT-64D(C) 0.834 0.368 0.388

5.4 Scale Change

The robustness of all proposed descriptors is veri-

ﬁed against scaling change by selecting 500 differ-

ent scenes of the benchmark database (Nist

er and

Stew

enius, 2006) for which there are two images

ICPRAM 2016 - International Conference on Pattern Recognition Applications and Methods

taken at different scales. Some of the selected images

have additional viewpoint change as well. The ﬁrst

image of each scene is used as a query image and the

second one is used as an available image in the image

database. Table 10 shows that both SIFT−128D and

RC-SIFT−64D perform consistent in the case of scale

change. Moreover, it presents that the SURF−64D

descriptors perform the worst in this case.

Table 10: Comparison of retrieval performance of

SIFT−128D, SIFT−64D, SURF−64D and our RC-SIFT-

64D using a ground truth database (500 images) and a set

of 500 query images, each of them has one salt and pep-

per noised image in the database. For each query image we

check if its corresponding database image appear as the ﬁrst

retrieved image in the result.

Method Recall

SIFT-128D 0.801

SIFT-64D 0.763

SURF-64D 0.533

RC-SIFT-64D(R) 0.801

RC-SIFT-64D(C) 0.801

5.5 Perspective Change

To test the invariance of descriptors against perspec-

tive change. 500 different scenes of the benchmark

database (Nist

er and Stew

enius, 2006) are selected for

which there are two images taken at different view-

point angles. The ﬁrst image of each scene is used

as a query image and the other one is used as an

image in the database. The results are presented in

Table 11, which describes that contrary to the other

kinds of changes, the robustness of all proposed de-

scriptors against perspective change decrease. But

SIFT−128D and RC-SIFT-64D still have the best per-

formance.

Table 11: Comparison of retrieval performance of

SIFT−128D, SIFT−64D, SURF−64D and our RC-SIFT-

64D using a ground truth database (500 images) and a set

of 500 query images, each of them has one salt and pep-

per noised image in the database. For each query image we

check if its corresponding database image appear as the ﬁrst

retrieved image in the result.

Method Recall

SIFT-128D 0.626

SIFT-64D 0.602

SURF-64D 0.439

RC-SIFT-64D(R) 0.720

RC-SIFT-64D(C) 0.717

6 CONCLUSION

In this work, we consider the fact that “the spar-

sity of ﬁxed amount of feature increase as their di-

mensionality increase” (Steinbach et al., 2003) to

reduce dimensionality of the SIFT descriptor from

128D to 64D. The goal of dimensionality reduc-

tion is to decrease the sparsity of SIFT descriptors,

speeding up the indexing process and improve the

performance of SIFT descriptors in the image NDR

ﬁeld. We veriﬁed in this work the performance

of the RC-SIFT−64D (for both horizontal and ver-

tical compression), RC-SIFT−32D, RC-SIFT−16D

against the original SIFT−128D to solve image NDR

tasks using a benchmark which contains different

kind of indoor/outdoor images. The experiments

show a slight improvement in matching results when

tested on benchmark databases. Moreover, the RC-

SIFT−64D needs shorter time for indexing and less

memory than the original SIFT−128D. However, the

performance of RC-SIFT−32D and RC-SIFT−16D

decrease, due to the compression of descriptors infor-

mation in both direction at once. The robustness and

stability of our suggested RC-SIFT−64D are veriﬁed

against different kinds of image afﬁne transformation,

blurring change, scale change and viewpoint change.

However, the results shows that the RC-SIFT−64D

descriptors are invariant to image afﬁne transforma-

tion in some speciﬁc ranges. Moreover, it presents

that RC-SIFT−64D descriptors are robust against im-

age blurring, scale change and perspective change. In

addition, its descriptors are more robustness than the

other presented descriptors against a big change in the

rotation, some kinds of noise, big amount of blurring

and viewpoint change. In the next step, we will at-

tempt to improve the performance of RC-SIFT−64D

by adapting suitable weights inspired by the rela-

tion between features to improve the performance of

matching in the ﬁeld of image NDR.

In future work, we will also evaluate if the more

robust performance of the RC-SIFT−64D can be used

in the ﬁeld of human visual attention, e.g., as a more

stable predictor for creating a saliency map of hu-

man gaze as discussed in a previous study (Steffen

et al., 2012). Furthermore, the more efﬁcient RC-

SIFT−64D approach may improve interactive image

search when a large scale image collection is used

as e.g., in (Low et al., 2014) and (Beecks and Seidl,

2009).

Adapted SIFT Descriptor for Improved Near Duplicate Retrieval

ACKNOWLEDGEMENTS

I would like to thank the state Saxony-Anhalt for the

ﬁnancial support of this work.

REFERENCES

Auclair, A., Vincent, N., and Cohen, L. (2006). Hash func-

tions for near duplicate image retrieval. In Applica-

tions of Computer Vision (WACV), pages 7–8.

Beecks, C. and Seidl, T. (2009). Visual exploration of large

multimedia databases. In Data Management and Vi-

sual Analytics Workshop.

Chu, L., Jiang, S., Wang, S., Zhang, Y., and Huang, Q.

(2013). Robust spatial consistency graph model for

partial duplicate image retrieval. In Multimedia, IEEE

Transactions on, pages 1982–1996.

Chum, O., Philbin, J., Isard, M., and Zisserman, A. (2007).

Scalable near identical image and shot detection. In

Proc. CIVR.

Chum, O., Philbin, J., and Zisserman, A. (2008). Near du-

plicate image detection: min-hash and tf-idf weight-

ing. In British Machine Vision Conference.

Grauman, K. and Darrell, T. (2005). Pyramid match ker-

nels: Discriminative classiﬁcation with sets of image

features. In Proc. ICCV.

Grauman, K. and Darrell, T. (2007). The pyramid match

kernel: Efﬁcient learning with sets of features. In The

Journal of Machine Learning Research, pages 725–

760.

egou, H., Douze, M., Schmid, C., and P

erez, P. (2010).

Aggregating local descriptors into a compact image

representation. In Proc. IEEE Conf. Computer Vision

and Pattern Recognition.

Jiang, M., Zhang, S., Li, H., and Metaxas, D. N. (2015).

Computer-aided diagnosis of mammographic masses

using scalable image retrieval. In Biomedical Engi-

neering, IEEE Transactions on, pages 783–792.

Ke, Y. and Sukthankar, R. (2004). Pca-sift: A more distinc-

tive representation for local image descriptors. In in:

CVPR, issue 2, page 506513.

Khan, N., McCane, B., and Wyvill, G. (2011). Sift and

surf performance evaluation against various image de-

formations on benchmark dataset. In Digital Image

Computing Techniques and Applications (DICTA).

Li, J., Qian, X., Li, Q., Zhao, Y., Wang, L., and Tang,

Y. Y. (2014). Mining near duplicate image groups.

In Springer Science and Business Media New York.

Low, T., Hentschel, C., Stober, S., Sack, H., and

urnberger, A. (2014). Visual berrypicking in large

image collections. In Proceedings of the 8th Nordic

Conference on Human-Computer Interaction: fun,

fast, foundational, pages 1043–1046. New York, NY :

ACM.

Lowe, D. (2004). Distinctive image features from scale-

invariant keypoints. In Journal of Computer Vision,

pages 91–110.

Nist

er, D. and Stew

enius, H. Recognition benchmark

images. In available at http://www.vis.uky.edu/

∼stewe/ukbench/.

Nist

er, D. and Stew

enius, H. (2006). Scalable recognition

with a vocabulary tree. In CVPR, pages 2161–2168.

Steffen, J., Christian, H., Alyosef, A. A., T

onnies, K., and

urnberger, A. (2012). Rotational invariance at ﬁxa-

tion points - experiments using human gaze data. In

Proceedings of the 1st International Conference on

Pattern Recognition Applications and Methods, pages

451–456.

Steinbach, M., Ertoz, L., and Kumar (2003). The challenges

of clustering high dimensional data. In Wille LT, ed-

itor. New Vistas in Statistical Physics-Applications in

Econophysics, Bioinformatics, and Pattern Recogni-

tion. Springer-Verlag.

Xu, D., Cham, T., Yan, S., Duan, L., and Chang, S. (2010).

Near duplicate identiﬁcation with spatially aligned

pyramid matching. In IEEE Trans. Circuits and Sys-

tems for Video Technology, pages 1068–1079.

Yang, Y. and Newsam, S. (2008). Comparing sift descrip-

tors and gabor texture features for classiﬁcation of re-

mote sensed imagery. In Proceedings of the 15th IEEE

on Image Processing, San Diego, pages 1852–1855.

USA.

Zhang, C., Wang, S., Huang, Q., Liu, J., Liang, C., and Tian,

Q. (2013). Image classiﬁcation using spatial pyramid

robust sparse coding. In Pattern Recognition letters,

pages 1046–1052.

Zhang, D. Q. and et al. (2004). Detecting image

near-duplicate by stochastic attribute relational graph

matching with learning. In Proceedings of the 12th

annual ACM international conference on Multimedia.

ICPRAM 2016 - International Conference on Pattern Recognition Applications and Methods