Robust Head-shoulder Detection using Deformable Part-based Models

Enes Dayangac, Christian Wiede, Julia Richter and Gangolf Hirtz

Faculty of Electrical Engineering and Information Technology, Technische Universit

at Chemnitz,

Chemnitz, Germany

Keywords:

Person Detection, Head-shoulder Detection, Ambient Assisted Living, Latent SVM, DPM, ACF-Detector.

Abstract:

Conventional person detection algorithms lack of robustness, especially when the person is partially occluded.

We propose thereby a robust head-shoulder detector in 2-D images using deformable part-based models. This

detector can be used in a variety of applications such as people counting and person dwell time measurements.

In experiments, we compare the head-shoulder detector with the full body detector quantitatively and analyze

the robustness of the detector in realistic scenarios. In the results, we show that the model learned with our

method outperforms other methods proposed in related work on an ambient assisted living application.

1 INTRODUCTION

Person detection ﬁnds application in a variety of ﬁelds

such as public security, automotive as well as in the

business sector. It furthermore plays a crucial role in

the upcoming application ﬁeld of Ambient Assisted

Living (AAL). The notion of AAL is to assist elderly

people during their daily life with the help of innova-

tive technology.

In our study, we have designed an AAL system

that applies optical sensors. One essential part of this

system is a machine learning-based person detection

algorithm. This paper focuses on the detection of per-

sons within a complex living environment. In this

context, a number of challenges have to be addressed.

Firstly, a person detection algorithm is required to

cope with varying illumination conditions and large

distances between persons and sensors. Secondly, in

contrast to many other studies, a wide-angle camera is

mounted on the ceiling of a room, so that a large area

of the scene can be covered. As a result, the person’s

lower part is occluded by the torso, especially when

the person is standing very close to the camera. In

typical AAL scenarios, such as sitting at a table, the

legs are also not visible either. However, the person

models used in state-of-the-art algorithms are trained

with samples of the full body from frontal view. Con-

sequently, these models show inadequate results and

are therefore not suitable for the presented applica-

tion.

For this reason a head-shoulder model for that spe-

cial camera set-up has been trained in the presented

study. As there are no appropriate public datasets

available, we recorded a variety of sequences under

different lighting and distance conditions. The per-

formance of the head-shoulder model was analyzed

using samples of our own dataset and samples from

a public dataset. We have chosen two state-of-the-

art algorithms that proved to work effectively with

full body models in front-view scenarios and modi-

ﬁed them by adapting the head-shoulder model.

2 RELATED WORK

In recent years the number of people detection al-

gorithms has been steadily increasing. In the past,

pedestrian detection was pursued by many researchers

in the context of Driver Assistance Systems. Doll

ar et

al. presented an overview about state-of-the-art algo-

rithms for pedestrian detection (Doll

ar et al., 2012).

They especially focussed on sliding window tech-

niques, which give promising results for low and

medium resolution images in which segmentation and

key point based approaches proved to be unsuitable.

One of the ﬁrst sliding window approaches applied

a support vector machine that was trained with Haar

Wavelets (Papageorgiou and Poggio, 2000). Building

upon this concept the approach of Viola and Jones in-

troduced integral images for fast feature calculation

(Viola and Jones, 2001). They used AdaBoost as

training method and a cascade structure for an efﬁ-

cient detection with a reduced number of false posi-

tives. This method serves as a basis for several mod-

236

Dayangac E., Wiede C., Richter J. and Hirtz G..

Robust Head-shoulder Detection using Deformable Part-based Models.

DOI: 10.5220/0005266002360243

In Proceedings of the 10th International Conference on Computer Vision Theory and Applications (VISAPP-2015), pages 236-243

ISBN: 978-989-758-090-1

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

ern detectors.

The development of advanced, highly-descriptive

feature types was essential with regard to the accu-

racy of detection algorithms. Especially gradient-

based features like Scale-Invariant Feature Transform

(SIFT) (Lowe, 2004) and Histogram of Oriented Gra-

dients (HOG) (Dalal and Triggs, 2005) played an im-

portant role in this context. Nowadays HOG features

are the most frequently used descriptors for person

detection. A considerable amount of studies, such as

(Zhu et al., 2006) and (Shashua et al., 2004), focused

on enhancing the detection results. Other studies in-

vestigated features that are based on shape (Gavrila,

2007), (Sabzmeydani and Mori, 2007) and motion in-

formation (Viola et al., 2003),(Wojek et al., 2009).

Further developments combined several different

features in order to get more powerful descriptors.

Since HOG is rated as one of the most effective sin-

gle feature, several studies investigated combinations

with HOG and other features. Wojek and Schiele

combined HOG, Haar-like features and shape features

(Wojek and Schiele, 2008). This approach was later

extended by color self similarity and motion informa-

tion (Walk et al., 2010).

Doll

ar et al. proposed another method (Doll

et al., 2009a), which takes up the idea of Viola and

Jones (Viola and Jones, 2001). They modiﬁed the

original approach by searching for Haar-like features

in different channels, such as the LUV color chan-

nel as well as gray-scale, gradient magnitude and ori-

entation images. In one of their following works,

they extended this approach to a multi-scale detec-

tion (Doll

ar et al., 2010). In this way processing time

was reduced because features can be computed from

a nearby scale. Another promising approach are part-

based models. In contrast to the previously described

approaches, the part-based models are constructed in

such a way that not only the person as a whole model

is considered, but different parts of the person are de-

scribed and used for classiﬁcation. By constructing

these part models, variances caused by rotation and

occlusion can be reduced. One of the algorithms pro-

viding the best results so far is the deformable part-

based models approach (DPM). It is a discrimina-

tive part-based approach, whereby unknown part po-

sitions are described as latent variables (Felzenszwalb

et al., 2008), (Felzenszwalb et al., 2010). As they uti-

lized a SVM for classiﬁcation, this classiﬁer is called

Latent SVM. This approach was later extended by

Park et al. to a multi-resolution model (Park et al.,

2010). Another approach that dealt with partial oc-

clusion was based on an Edgelet detector, which is a

kind of shape detector combined with AdaBoost (Wu

and Nevatia, 2005).

Beside algorithms processing single, monocular

images, there is furthermore a variety of 3-D based

approaches. Kirchner et al. segment a person’s point

cloud into horizontal slices (Kirchner et al., 2012).

The span of each slice is accumulated in a feature

vector that is used for training a SVM. Richter et al.

localize persons in 3-D world coordinates by ﬁrstly

generating foreground hypotheses on the world z-map

and then projecting 3-D points onto a virtual overhead

view (Richter et al., 2014). In this study, the authors

make the assumption that foreground objects of a cer-

tain size are most probable persons. However, it could

be sensible to validate the detected persons by means

of a 2-D based person detection algorithm.

The latest 2-D person detection algorithms utilize

models that are composed of the full body. Since

the camera is monitoring the scene at a certain an-

gle with respect to the ceiling in many AAL appli-

cations (side and top view if the person is standing

near the camera), the full body model is not suitable.

In general, head-shoulder detection is more reliable

than the full-body detection even in highly occluded

cases (Tu et al., 2013). Therefore a new, more efﬁ-

cient model has to be trained. As the shoulder part

is always visible in this camera set-up, it is rational

to train a model with samples showing the person’s

shoulder part. There already exist several approaches

for head-shoulder detection. Li et al. proposed a

method which detects the head-shoulder part by com-

bining a Viola-Jones type classiﬁer and a local HOG

feature-based AdaBoost classiﬁer (Li et al., 2009).

They furthermore showed how to track head-shoulder

parts by a particle ﬁlter approach. An attention-based

foreground segmentation was combined with a multi-

view cascade to detect head-shoulder parts for video

surveillance (Tu et al., 2013). Wang et al. introduced

a new edge feature called En-Contour (Wang et al.,

2013). However, these mentioned methods have not

been studied and tested in AAL scenarios.

At present there is no public head-shoulder

database available that can be used for training.

Commonly-known datasets such as INRIA (Dalal and

Triggs, 2005), PASCAL (Everingham et al., 2010),

Caltech (Doll

ar et al., 2009b), ETH (Ess et al., 2007)

and Daimler (Enzweiler and Gavrila, 2009) only pro-

vide sequences where the camera is installed at ap-

proximately the same height as the recorded persons.

Therefore the whole body is always visible and the

optical axis is almost perpendicular to the person’s

main axis. For this reason we recorded an own dataset

for both training and testing purposes.

RobustHead-shoulderDetectionusingDeformablePart-basedModels

237

3 METHODS

Deformable part-based models (Felzenszwalb et al.,

2010) can overcome major challenges arising in ob-

ject detection. Effects such as deformation, occlusion

and viewpoint changes can be managed to a certain

extent. However, in many cases in AAL scenarios

the described effects reach such a degree that persons

cannot be detected any more when applying the com-

monly used full body models. Examples of such sce-

narios are illustrated in Fig. 1.

Figure 1: Person detection when using the full body model:

Persons standing very close or almost below the camera,

occluded by objects such as a table, or with the back turned

to the camera cannot be reliably detected. There often occur

false positives or false negatives in such cases.

Nevertheless it can be observed that in all those

images the head-shoulder part is always visible. For

this reason, we propose to use this very part to build a

new model for the DPM classiﬁer. By using this La-

tent SVM based approach, a dynamical assignment

of part models to overall models is possible. Since

we propose to use the DPM approach with consider-

ing detection of the head-shoulder of a person, this

method is explained in this section.

The model is characterized by a lower-resolution

root ﬁlter, several spatially ﬂexible, higher-resolution

part ﬁlters and a spatial model for the locations of ev-

ery part with respect to the root. The generated model

is shown in Fig. 2.

Figure 2: Illustration of a head-shoulder model. The left

image shows the lower-resolution root ﬁlter, the image in

the middle demonstrates higher-resolution part ﬁlters and

the right image the spatial model.

When a person shall be localized, a feature pyra-

mid is created in the ﬁrst step by down-scaling the

image. After the calculation of HOG features at every

scale, the ﬁlters are applied in order to get the ﬁlter re-

sponse for every single part model. The part models

are then combined at one scale and the results of ev-

ery scale are ﬁnally fused in order to compute a ﬁnal

score for the root locations. Fig. 3 shows the result

of the example image (see second image in Fig. 1),

where the locations of the found parts and the root are

marked by blue and red rectangles respectively.

Figure 3: Exemplary detection of the DPM algorithm using

the proposed head-shoulder model. When using the head-

shoulder model, the person sitting at the table can be de-

tected.

4 DATA PROCESSING CHAIN

In order to get a powerful person model, several steps

starting from data acquisition to model evaluation

were performed.

4.1 Data Acquisition

For training purposes, a CCD Camera (Allied

GC1350) with 1.3 MP and a wide lens were used to

cover the whole room. In some scenarios, the camera

is mounted at the ceiling with a tilt angle of about 20

and a height of 2 m from the ﬂoor. In other scenarios,

the camera is held at a certain height in front of the

persons. Generally it is important to build a model

in such a way that it shows good performance under

different conditions. Therefore the exposure time as

well as the gain of the camera were set to different

values in order to obtain images recorded under vary-

ing lighting conditions.

Altogether, 11920 images (9013 positive and 2907

negative) were collected whereas 13 male and female

persons of different age groups participated in the

recordings. All persons wore different clothing to get

variations in our dataset. In the scenarios, the actors

were sitting, standing and lying.

The negative images were randomly collected

from the Internet. In these images no persons are

present.

VISAPP2015-InternationalConferenceonComputerVisionTheoryandApplications

238

4.2 Data Annotation and Database

For data annotation, ﬁles in PASCAL style (Evering-

ham et al., 2010) were generated from our data col-

lection. In order to annotate an object in an image,

we speciﬁed bounding boxes and labelling conditions

such as segmented, truncated, occluded, or difﬁcult.

Multiple objects from multiple classes may be present

in the same image. However, in our dataset, only one

object was allowed in the same image.

Therefore, positive images each contain exactly

one single person. Every person’s head-shoulder part

was speciﬁed by manually ﬁtting a bounding box to

this region in the image. As this process was very

time consuming, a labelling tool was developed to in-

crease the annotation speed. It enabled the drawing of

the bounding rectangles in the images intuitively.

The annotated data was split into training and test-

ing data sets. For a sensible evaluation, these two

datasets were independent from each other.

4.3 Training

A few models were trained to detect both full body

and head-shoulder using the DPM and ACF detector.

Both training and testing were performed in Mat-

lab code. The Matlab Toolbox ”Piotr’s Image and

Video Matlab Toolbox” (Doll

ar, 2013) and an ob-

ject detection system using deformable part mod-

els (DPMs) and latent SVM (voc-release5) (Girshick

et al., 2012) (Felzenszwalb et al., 2010) are utilized.

5 EXPERIMENTAL SETUP AND

RESULTS

For evaluation, a number of experiments were per-

formed. Firstly, two algorithms were compared while

considering the head-shoulder and the full body mod-

els. Secondly, we analyzed the performance of the

head-shoulder detector in our approach by tuning sev-

eral parameters. Finally, the performance of the pro-

posed detector was evaluated by a public dataset. Ta-

ble 1 gives an overview about the experiments.

Precision-recall curves were used to compare the

two considered algorithms. False positive rate (FPR)

as well as true positive rate (TPR) were determined

at different threshold values. If the overlap ratio be-

tween the detected and the labelled rectangle was

higher than a certain value the sample is counted as

true positive else it is a false positive.

Table 1: Experimental overview.

Experiments Content

Experiment 1 DPM vs. ACF detector

full body

(our dataset)

Experiment 2 DPM vs. ACF detector

head-shoulder

(our dataset)

Experiment 3 head-shoulder vs. full body (DPM)

human pose (sitting, standing)

partial occlusion

(our dataset)

Experiment 4 head-shoulder (DPM)

distance (near, mid, far)

lighting (bright, dark)

(our dataset)

Experiment 5 head-shoulder (DPM)

number of images used for training

(our dataset)

Experiment 6 head-shoulder (DPM)

(public dataset)

Figure 4: The performance of the full body DPM model

(trained using the Pascal dataset, cascade PCA 5, the over-

lap area > 10%).

5.1 Experiment 1

In the ﬁrst experiment, we compared the full body

DPM and ACF detector deploying a set of test data

shown in Table 3.

However, these two detectors have been trained:

The DPM was trained with the Pascal person dataset

while ACF detector was trained with the Inria dataset.

The performances of the full body DPM and ACF de-

tector on the testing data are shown in Fig. 4 and Fig.

5 respectively.

RobustHead-shoulderDetectionusingDeformablePart-basedModels

239

Figure 5: The performance of the full body ACF detector

(trained using the Inria dataset, the overlap area > 10%).

5.2 Experiment 2

In this experiment, we trained the models for detect-

ing the head-shoulder part using our own dataset de-

signed in Sect 4. The purpose is to analyze the per-

formance of two different algorithms using the same

training and testing data for the head-shoulder. De-

ploying the testing data shown in Table 3, the perfor-

mances of the models are shown in Fig. 6 and Fig.

Deploying the ACF detector, different parame-

ters’ values are assigned to train several ACF detector

models.

5.3 Experiment 3

In this experiment, the full body model was compared

against the head-shoulder model. Both have been

trained with the DPM approach. The test data used

in this experiment consists of 1601 images with three

Figure 6: The performance of head-shoulder DPM model

(trained using around 9000 positive images, cascade PCA

5, the overlap area > 50%).

different persons wearing different clothing. They

performed actions of sitting (on a chair, armchair and

bed), standing or walking. In this scenario, the per-

sons were often partially occluded and not always

fully visible. The TPR and FPR results are shown

in Table 2.

Table 2: Comparison of the full body and the head-shoulder

model.

Model TPR FPR

Full body 48 % 1 %

Head-shoulder 93 % 2 %

5.4 Experiment 4

For this experimental setup, we considered two pa-

rameters: distance and lighting conditions. There are

three distance values (near, mid and far), which de-

scribe the distance between a person and the camera.

The lighting can be either dark or bright, see Fig. 8.

Table 3: Number of images for the different conﬁgurations.

Lighting, Distance Number of Images

Dark-near 309

Dark-mid 306

Dark-far 338

Bright-near 301

Bright-mid 360

Bright-far 305

Consequently, this results in six different param-

eter conﬁgurations, see Table 3. The number of im-

ages used in every conﬁguration is almost equally dis-

tributed.

In Fig. 6, the precision and recall curves illustrate

the performance results including average precision

(AP).

Figure 7: The performance of head-shoulder ACF detector

(trained using around 9000 positive images, the overlap area

> 50%).

VISAPP2015-InternationalConferenceonComputerVisionTheoryandApplications

240

Figure 8: The left and right images are the samples in the

test group of dark-mid and bright-near respectively.

5.5 Experiment 5

In this experiment, 9000 positive images were

used. Furthermore, another DPM head-shoulder was

trained using 1000 positive images for the purpose

of comparing the effect of using different number of

training data. Fig. ﬁg:exp2-1 and Fig. 9 presents the

performance of these models respectively on the test

data shown in Table 3.

Figure 9: The performance of head-shoulder DPM model

(trained using around 1000 positive images, cascade PCA

5, the overlap area > 50%).

5.6 Experiment 6

In the last experiment, we evaluated the performance

of head-shoulder DPM model on a public dataset. In

that way, we tested whether our trained models are

still effective in other AAL scenarios as well. There-

fore, the test data set from GER’HOME laboratory

(Zouba et al., 2007) was employed. The result is

shown in Table 4.

Table 4: Test on public data set.

Images Model TPR FPR

GER’HOME Full body 76 % 12 %

GER’HOME Head-shoulder 93 % 2 %

6 DISCUSSION

6.1 DPM vs. ACF Detector

In the ﬁrst and second experiments (see the compari-

sion of Fig. 4, 5 as well as see Fig. 6, 7), the results

show that the DPM models are better than ACF de-

tector for both full body and head-shoulder models.

With regard to the experiments comparing the re-

sults of dark far views with bright far views in Fig. 5

and Fig. 7, one can say that the ACF detector fails

to detect a person in bright images. It might be due to

the different training dataset that was used to train that

model. In the ACF detector approach, the distance

factor does not play any role for the performance of

the model.

6.2 Head-shoulder vs. Full Body

The results in Table 2 show that the TPR of the head-

shoulder model is two times higher than the TPR of

the full body model. When an appropriate thresh-

old is applied to the detection scores, the DPM head-

shoulder works more robustly than the DPM full body

model.

The head-shoulder model gives very good results

in the following scenarios: parts of the person are oc-

cluded while the person is sitting, the person is stand-

ing behind other objects and the person is standing

very close to the camera. In these cases, the full

body model shows poor performance. Furthermore,

the full body model often fails to detect the person

while standing up, sitting down and bending to pick

up an object from on the ﬂoor.

Moreover, the head-shoulder model is tested on

public datasets and it is proven that the head-shoulder

detection has higher detection rates than the full body

model (see Table 2 and 4). By means of these experi-

ments it could be demonstrated that the head-shoulder

model is more efﬁcient in AAL scenarios than the full

body model.

6.3 Distance

In Sect. 5.4, the distance factor in the model perfor-

mance was evaluated. Distance refers to the persons’

positions with respect to the camera. One purpose of

the study is to design a person detection model which

is independent from the distance.

In Fig. 6 and Fig. 9, the results show that the

distance factor has no inﬂuence on the head-shoulder

model. For instance, even if the person is near, the

head-shoulder is still visible and the head-shoulder

silhouette is not affected by the perspective distortion.

RobustHead-shoulderDetectionusingDeformablePart-basedModels

241

However, the distance factor has effects on the

full body detection. In our observation, the full body

model has good precision and higher recall when the

person is far, which means that the person’s full body

is in the view. When the person is near, the full body

model has a very low recall due to top view, huge per-

spective distortion and occlusion of the lower body.

6.4 Lighting

In Fig. 6 and Fig. 9, the precision-recall curve shows

that the DPM head-shoulder model performs better

under brighter light conditions. In Fig. 4, the recall

value reaches 0.8 when the images have a high con-

trast and the person is far from the camera. However,

this value reaches can only 6.2 in the low contrast im-

ages. To sum up, we can say that generally, good il-

lumination condition yields a better result without de-

pending on the distance of the person to the camera.

Nonetheless, we believe that the performance rel-

atively depends on the training dataset. If we would

have trained a model using more number of low-

contrast images, the result might be reversal.

6.5 Number of Training Samples

Two head-shoulder models are trained using a differ-

ent number of positives. In Sect. 5.5, one part of the

experiment is to compare these models. The model

with a lower number of positive images (see Fig. 9)

has higher precision rates but less recall compared to

the model trained with more number of positive im-

ages (see Fig. 6). Thus, training a model with more

positive images generalizes the parameters. There-

fore, more false detection alarms occur when lower-

ing the precision rates (see Fig. 9).

7 CONCLUSIONS

Person detection plays an important role in many ap-

plications. For AAL applications, we analyzed the

performance of state-of-the art person detection algo-

rithms using full body models. It was proven that they

lack robustness especially when parts of the person

are occluded, e. g. because the person is standing very

close to a tilted camera mounted at the ceiling or if the

person is turned with the back to the camera while sit-

ting. We therefore introduced a head-shoulder model,

because this part is visible in most cases.

For training and testing purposes, we collected our

own data set and annotated it. The performance was

analyzed for the DPM and the ACF detector. Further-

more, we compared the efﬁciency of the head shoul-

der model and the full body model. The results show

that the head-shoulder model is more robust than the

full body model in AAL scenarios. Another ﬁnd-

ing was that the DPM outperforms the ACF detector.

We proved on a public dataset that the head-shoulder

DPM model is very efﬁcient as well. In addition to

that, the detector was successfully tested under differ-

ent distance and lighting conditions.

In future, the DPM head-shoulder model has to be

enhanced with further training samples showing other

view points of persons, e. g. lying in the bed. Natu-

rally, this approach could also ﬁnd usage in a wide

range of other application ﬁelds, like security, con-

sumer market or public facilities.

ACKNOWLEDGMENTS

This study is funded by the European Fund for Re-

gional Development (EFRE). Special thanks are fur-

thermore expressed to all persons participating in the

recording and labelling process.

REFERENCES

Dalal, N. and Triggs, B. (2005). Histograms of oriented gra-

dients for human detection. In Computer Vision and

Pattern Recognition, 2005. CVPR 2005. IEEE Com-

puter Society Conference on, volume 1, pages 886–

893 vol. 1.

Doll

ar, P. (2013). Piotr’s Image and Video Matlab Toolbox

(PMT). http://vision.ucsd.edu/∼pdollar/toolbox/doc/

index.html.

Doll

ar, P., Belongie, S., and Perona, P. (2010). The fastest

pedestrian detector in the west. In BMVC.

Doll

ar, P., Tu, Z., Perona, P., and Belongie, S. (2009a). In-

tegral channel features. In BMVC.

Doll

ar, P., Wojek, C., Schiele, B., and Perona, P. (2009b).

Pedestrian detection: A benchmark. In CVPR.

Doll

ar, P., Wojek, C., Schiele, B., and Perona, P. (2012).

Pedestrian detection: An evaluation of the state of the

art. PAMI, 34.

Enzweiler, M. and Gavrila, D. (2009). Monocular pedes-

trian detection: Survey and experiments. Pattern

Analysis and Machine Intelligence, IEEE Transac-

tions on, 31(12):2179–2195.

Ess, A., Leibe, B., and Van Gool, L. (2007). Depth and ap-

pearance for mobile scene analysis. In Computer Vi-

sion, 2007. ICCV 2007. IEEE 11th International Con-

ference on, pages 1–8.

Everingham, M., Van Gool, L., Williams, C., Winn, J., and

Zisserman, A. (2010). Pascal. International Journal

of Computer Vision, 88(2).

Felzenszwalb, P., Girshick, R., McAllester, D., and Ra-

manan, D. (2010). Object detection with discrim-

VISAPP2015-InternationalConferenceonComputerVisionTheoryandApplications

242

inatively trained part-based models. Pattern Analy-

sis and Machine Intelligence, IEEE Transactions on,

32(9):1627–1645.

Felzenszwalb, P., McAllester, D., and Ramanan, D. (2008).

A discriminatively trained, multiscale, deformable

part model. In Computer Vision and Pattern Recog-

nition, 2008. CVPR 2008. IEEE Conference on, pages

1–8.

Gavrila, D. (2007). A bayesian, exemplar-based ap-

proach to hierarchical shape matching. Pattern Anal-

ysis and Machine Intelligence, IEEE Transactions on,

29(8):1408–1421.

Girshick, R. B., Felzenszwalb, P. F., and

McAllester, D. (2012). Discriminatively

trained deformable part models, release 5.

http://people.cs.uchicago.edu/ rbg/latent-release5/.

Kirchner, N., Alempijevic, A., and Virgona, A. (2012).

Head-to-shoulder signature for person recognition. In

Robotics and Automation (ICRA), 2012 IEEE Interna-

tional Conference on, pages 1226–1231.

Li, M., Zhang, Z., Huang, K., and Tan, T. (2009). Rapid and

robust human detection and tracking based on omega-

shape features. In Image Processing (ICIP), 2009 16th

IEEE International Conference on, pages 2545–2548.

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

invariant keypoints. International Journal of Com-

puter Vision, 60:91–110.

Papageorgiou, C. and Poggio, T. (2000). A trainable system

for object detection. International Journal of Com-

puter Vision, 38(1):15–33.

Park, D., Ramanan, D., and Fowlkes, C. (2010). Multires-

olution models for object detection. In Computer Vi-

sion ECCV 2010, volume 6314 of Lecture Notes in

Computer Science, pages 241–254.

Richter, J., Findeisen, M., and Hirtz, G. (2014). Assess-

ment and Care System Based on People Detection

for Elderly Suffering From Dementia. In Consumer

Electronics Berlin (ICCE-Berlin), 2014. ICCEBerlin

2014. IEEE Fourth International Conference on Con-

sumer Electronics, pages 59–63. IEEE.

Sabzmeydani, P. and Mori, G. (2007). Detecting pedestri-

ans by learning shapelet features. In Computer Vision

and Pattern Recognition, 2007. CVPR ’07. IEEE Con-

ference on, pages 1–8.

Shashua, A., Gdalyahu, Y., and Hayun, G. (2004). Pedes-

trian detection for driving assistance systems: single-

frame classiﬁcation and system level performance. In

Intelligent Vehicles Symposium, 2004 IEEE, pages 1–

Tu, J., Zhang, C., and Hao, P. (2013). Robust real-time

attention-based head-shoulder detection for video

surveillance. In Image Processing (ICIP), 2013 20th

IEEE International Conference on, pages 3340–3344.

Viola, P. and Jones, M. (2001). Robust real-time object de-

tection. In International Journal of Computer Vision.

Viola, P., Jones, M., and Snow, D. (2003). Detecting pedes-

trians using patterns of motion and appearance. In

Computer Vision, 2003. Proceedings. Ninth IEEE In-

ternational Conference on, pages 734–741 vol.2.

Walk, S., Majer, N., Schindler, K., and Schiele, B. (2010).

New features and insights for pedestrian detection.

In Computer Vision and Pattern Recognition (CVPR),

2010 IEEE Conference on, pages 1030–1037.

Wang, S., Zhang, J., and Miao, Z. (2013). A new edge fea-

ture for head-shoulder detection. In Image Processing

(ICIP), 2013 20th IEEE International Conference on,

pages 2822–2826.

Wojek, C. and Schiele, B. (2008). A performance evalua-

tion of single and multi-feature people detection. In

Pattern Recognition, volume 5096 of Lecture Notes in

Computer Science, pages 82–91.

Wojek, C., Walk, S., and Schiele, B. (2009). Multi-cue on-

board pedestrian detection. In Computer Vision and

Pattern Recognition, 2009. CVPR 2009. IEEE Con-

ference on, pages 794–801.

Wu, B. and Nevatia, R. (2005). Detection of multiple, par-

tially occluded humans in a single image by bayesian

combination of edgelet part detectors. In Computer

Vision, 2005. ICCV 2005. Tenth IEEE International

Conference on, volume 1, pages 90–97 Vol. 1.

Zhu, Q., Yeh, M.-C., Cheng, K.-T., and Avidan, S. (2006).

Fast human detection using a cascade of histograms

of oriented gradients. In Computer Vision and Pat-

tern Recognition, 2006 IEEE Computer Society Con-

ference on, volume 2, pages 1491–1498.

Zouba, N., Bremond, F., Thonnat, M., and Vu, V. T. (2007).

Multi-sensors analysis for everyday activity monitor-

ing. Proc. of SETIT, pages 25–29.

RobustHead-shoulderDetectionusingDeformablePart-basedModels

243