Active Shape Model Research Papers (original) (raw)

Abstract. In this paper we introduce the concept of statistical deformation mod-els (SDM) which allow the construction of average models of the anatomy and their variability. SDMs are built by performing a statistical analysis of the... more

Abstract. In this paper we introduce the concept of statistical deformation mod-els (SDM) which allow the construction of average models of the anatomy and their variability. SDMs are built by performing a statistical analysis of the defor-mations required to map anatomical features in one subject into the corresponding features in another subject. The concept of SDMs is similar to active shape mod-els (ASM) which capture statistical information about shapes across a population but offers several new advantages over ASMs: Firstly, SDMs can be constructed directly from images such as MR or CT without the need for segmentation which is usually a prerequisite for the construction of active shape models. Instead a non-rigid registration algorithm is used to compute the deformations required to establish correspondences between the reference subject and the subjects in the population class under investigation. Secondly, SDMs allow the construction of an atlas of the average anatomy as we...

The aim of this project was to investigate the nature and possible significance of first-person kinaesthetic vocal sensations observed in association with musical listening. Hearing and voice are known to be closely linked but the... more

The aim of this project was to investigate the nature and possible significance of first-person kinaesthetic vocal sensations observed in association with musical listening. Hearing and voice are known to be closely linked but the mechanisms that underlie their close relationship are not yet understood. The presence of kinaesthetic vocal sensations challenges accounts of auditory processing that are divorced from peripheral vocal input and, instead, suggests the hypothesis that auditory and vocal processing mechanisms rely on shared peripheral substrates in addition to their increasingly recognized shared (brain-based) central substrates. To investigate this hypothesis, I used MRI and developed a measurement protocol (informed by established methods in cephalometry) that would allow me to relate vocal structures to their direct and indirect bony attachments to the craniofacial skeleton, cervical spine and sternum. After establishing the method’s validity in subjects at rest, I acquired midsagittal MR images (under conditions where articulatory and postural input was negligible) while subjects (1) hummed and (2) listened (in a focused way) to low and high notes at each end of their range. Geometric and shape analysis of craniocaudal, craniocervical and anteroposterior variables revealed significant differences between low- and high-note conditions and widespread correlations between variables for both humming and listening investigations. An unexpected association between pitch change and changes of cervical alignment was also found. These results were complemented and extended by using the same MR images to build an active shape model (ASM). In addition to showing how vocal structures move together, ASM showed goal-related vocal activity to consist of one or more independent modes of variation. Together, the observations, experimental results, and evidence from diverse historical and contemporary sources, support the hypothesis that mechanisms underlying auditory and vocal processing rely on shared central and peripheral substrates. Wide-ranging implications arising from this hypothesis are also discussed.

Recent advances in 3D scanning technology have enabled the development of interesting applications of 3D human body modelling and shape analysis, especially in the areas of virtual shopping, custom clothing and sizing surveys for the... more

Recent advances in 3D scanning technology have enabled the development of interesting applications of 3D human body modelling and shape analysis, especially in the areas of virtual shopping, custom clothing and sizing surveys for the clothing industry. Most of the current applications have so far been concerned with automatic tape measurement extraction, i.e. simulation of the manual procedure for extracting

Statistical shape analysis techniques commonly employed in the medical imaging community, such as active shape models or active appearance models, rely on principal component analysis (PCA) to decompose shape variability into a reduced... more

Statistical shape analysis techniques commonly employed in the medical imaging community, such as active shape models or active appearance models, rely on principal component analysis (PCA) to decompose shape variability into a reduced set of interpretable components. In this paper we propose principal factor analysis (PFA) as an alternative and complementary tool to PCA providing a decomposition into modes of variation that can be more easily interpretable, while still being a linear efficient technique that performs dimensionality reduction (as opposed to independent component analysis, ICA). The key difference between PFA and PCA is that PFA models covariance between variables, rather than the total variance in the data. The added value of PFA is illustrated on 2D landmark data of corpora callosa outlines. Then, a study of the 3D shape variability of the human left femur is performed. Finally, we report results on vector-valued 3D deformation fields resulting from non-rigid registration of ventricles in MRI of the brain.

We present a gait recognition system using infra-red (IR) images. Since an IR camera is not affected by the intensity of illumination, it is able to provide constant recognition performance regardless of the amount of illumination.... more

We present a gait recognition system using infra-red (IR) images. Since an IR camera is not affected by the intensity of illumination, it is able to provide constant recognition performance regardless of the amount of illumination. Model-based object tracking algorithms enable robust tracking with partial occlusions or dynamic illumination. However, this algorithm often fails in tracking objects if strong edge exists near the object. Replacement of the input image by an IR image guarantees robust object region extraction because background edges do not affect the IR image. In conclusion, the proposed gait recognition algorithm improves accuracy in object extraction by using IR images and the improvements finally increase the recognition rate of gaits.

The detection of double edges in x-ray images of lumbar vertebrae is of prime importance in the assessment of vertebral injury or collapse that may be caused by osteoporosis and other spine pathology. In addition, if the above double-edge... more

The detection of double edges in x-ray images of lumbar vertebrae is of prime importance in the assessment of vertebral injury or collapse that may be caused by osteoporosis and other spine pathology. In addition, if the above double-edge detection process is conducted within an automatic framework, it would not only facilitate inexpensive and fast means of obtaining objective morphometric measurements on the spine, but also remove the human subjectivity involved in the morphometric analysis. This paper proposes a novel force-formulation scheme, termed as Pressurized Open Directional Gradient Vector Flow snakes, to discriminate and detect the superior and inferior double edges present in the radiographic images of the lumbar vertebrae. As part of the validation process, this algorithm is applied to a set of 100 lumbar images and the detection results are quantified using analyst-generated ground truth. The promising nature of the detection results bears testimony to the efficacy of the proposed approach

This paper describes a novel idea of face recognition across facial expression variations using model based approach. The approach follows in 1) modeling an active appearance model (AAM) for the face image, 2) using optical flow based... more

This paper describes a novel idea of face recognition across facial expression variations using model based approach. The approach follows in 1) modeling an active appearance model (AAM) for the face image, 2) using optical flow based temporal features for facial expression variations estimation, 3) and finally applying binary decision trees as a classifier for facial identification. The novelty lies not only in generation of appearance models which is obtained by fitting active shape model (ASM) to the face image using objective but also using a feature vector which is the combination of shape, texture and temporal parameters that is robust against facial expression variations. Experiments have been performed on Cohn-Kanade facial expression database using 61 subjects of the database with image sequences consisting of more than 4000 images. This achieved successful recognition rate up to 91.17% using decision tree as classifier in the presence of six different facial expressions.

This paper explains how the shape of the spine can be evaluated from back surface measurements in a recumbent position, by using point distribution models (PDM) and typical shape variability of the spine in a lateral sleeping position.... more

This paper explains how the shape of the spine can be evaluated from back surface measurements in a recumbent position, by using point distribution models (PDM) and typical shape variability of the spine in a lateral sleeping position. CT-scans of 12 volunteers were taken in this posture on a firm and a soft sleeping system to provide a training set for the PDM. Active shape models (ASM) were used to enhance the accuracy of the spinal reconstruction from measurements by limiting the shape of the spine to characteristic shapes from a biomechanical and/or clinical point of view. A comparison was made between calculated shapes, obtained from surface measurements, and those measured vertebral body centres (from CT-scans). An RMS accuracy of 2.6mm was obtained in 3D, and 1.8mm in frontal view, which was sufficient to compare spinal deformations of a subject on different sleeping systems.

We describe the use of flexible models for representing the shape and grey-level appearance of human faces. These models are controlled by a small number of parameters which can be used to code the overall appearance of a face for image... more

We describe the use of flexible models for representing the shape and grey-level appearance of human faces. These models are controlled by a small number of parameters which can be used to code the overall appearance of a face for image compression and classification ...

In this work we propose a fully automatic system for bone age evaluation, according to the Tanner and Whitehouse method (TW2), based on the integration between EMROI and CROI analysis, which ensures accurate bone age assessment for the... more

In this work we propose a fully automatic system for bone age evaluation, according to the Tanner and Whitehouse method (TW2), based on the integration between EMROI and CROI analysis, which ensures accurate bone age assessment for the entire age range (0-10). For both approaches novel segmentation techniques will be proposed. In detail, for the CROI analysis the bones extraction has been carried out by integrating anatomical knowledge of the hand and trigonometric concepts, whereas the TW2 stage assignment is implemented by combining the active contour models (ACM) and derivative difference of Gaussian (DrDoG) filter. For the EMROI analysis, image processing techniques and geometrical features analysis, based on difference of Gaussian (DoG), are proposed. The experimental results were conducted on a set of 30 X-Rays, reaching performances of about 87%. The performances of the proposed method are affected by the detection and extraction of the Trapezium and Trapezoid (50%). Without considering such bones the success rate raises to 91%.

The detection of pedestrians in real-world scenes is a daunting task, especially in crowded situations. Our experience over the last years has shown that active shape models (ASM) can contribute significantly to a robust pedestrian... more

The detection of pedestrians in real-world scenes is a daunting task, especially in crowded situations. Our experience over the last years has shown that active shape models (ASM) can contribute significantly to a robust pedestrian detection system. The paper starts with an overview of shape model approaches, it then explains our approach which builds on top of Eigenshape models which have been trained using real-world data. These models are placed over candidate regions and matched to image gradients using a scoring function which integrates i) point distribution, ii) local gradient orientations iii) local image gradient strengths. A matching and shape model update process is iteratively applied in order to fit the flexible models to the local image content. The weights of the scoring function have a significant impact on the ASM performance. We analyze different settings of scoring weights for gradient magnitude, relative orientation differences, distance between model and gradient in an experiment which uses real-world data. Although for only one pedestrian model in an image computation time is low, the number of necessary processing cycles which is needed to track many people in crowded scenes can become the bottleneck in a real-time application. We describe the measures which have been taken in order to improve the speed of the ASM implementation and make it real-time capable.

This paper introduces a method for automatic facial expression recognition in image sequences, which make use of Candide wire frame model and active appearance algorithm for tracking, and support vector machine for classification. Candide... more

This paper introduces a method for automatic facial expression recognition in image sequences, which make use of Candide wire frame model and active appearance algorithm for tracking, and support vector machine for classification. Candide wire frame model is adapted properly on the first frame of face image sequence. Facial features in subsequent frames of image sequence are tracked using active

Facial Expression Recognition is a hot topic in recent years. As artificial intelligent technology is growing rapidly, to communicate with machines, facial expression recognition is essential. The recent feature extraction methods for... more

Facial Expression Recognition is a hot topic in recent years. As artificial intelligent technology is growing rapidly, to communicate with machines, facial expression recognition is essential. The recent feature extraction methods for facial expression recognition are similar to face recognition, and those caused heavy load for calculation. In this paper, Digitalized Facial Features based on Active Shape Model method is used to reduce the computational complexity and extract the most useful information from the facial image. The result shows by using this method the computational complexity is dramatically reduced, and very good performance was obtained compared with other extraction methods.