Abstract Aim: The aim was to evaluate height measurements made with the single view metrology method and to investigate the influence of standing position and different phases of gait and running on vertical height. Method: Ten healthy men were recorded simultaneously by a 2D web camera and a 3D motion analysis system. They performed six trials, three standing and three during gait and running. The vertical height was measured with the single view metrology method and in Qualisys Track Manager. The results were compared for evaluation. The vertical height in the different postures was compared to the actual height. Results: The measurements made with the single view metrology method were significantly higher than the measurements made with Qualisys Track Manager (p<0.001). The vertical height in the two standing positions was significantly lower than the actual height (p<0.05). The vertical height in midstance was significantly lower than actual height in the walking trials (p<0.05). No significant difference was found between maximum vertical height and actual height during running (p>0.05). Conclusion: The single view metrology method measured vertical heights with a mean error of +2.30 cm. Posture influence vertical body height. Midstance in walking is the position where vertical height corresponds best with actual height, in running it is the non-support phase.

<p><p><strong>Abstract</strong></p><p><strong>Aim: </strong>The aim was to evaluate height measurements made with the single view metrology method and to investigate the influence of standing position and different phases of gait and running on vertical height.</p><p><strong>Method: </strong>Ten healthy men were recorded simultaneously by a 2D web camera and a 3D motion analysis system. They performed six trials, three standing and three during gait and running. The vertical height was measured with the single view metrology method and in Qualisys Track Manager. The results were compared for evaluation. The vertical height in the different postures was compared to the actual height.</p><p><strong>Results: </strong>The measurements made with the single view metrology method were significantly higher than the measurements made with Qualisys Track Manager (p<0.001). The vertical height in the two standing positions was significantly lower than the actual height (p<0.05). The vertical height in midstance was significantly lower than actual height in the walking trials (p<0.05). No significant difference was found between maximum vertical height and actual height during running (p>0.05).</p><p><strong>Conclusion: </strong>The single view metrology method measured vertical heights with a mean error of +2.30 cm. Posture influence vertical body height. Midstance in walking is the position where vertical height corresponds best with actual height, in running it is the non-support phase.</p><p> </p></p><p> </p>

Recovery of human pose from videos has become a highly active research area in the last decade because of many attractive potential applications, such as surveillance, non-intrusive motion analysis and natural human machine interaction. Video based full body pose estimation is a very challenging task, because of the high degree of articulation of the human body, the large variety of possible human motions, and the diversity of human appearances. Methods for tackling this problem can be roughly categorized as either discriminative or generative. Discriminative methods can work on single images, and are able to recover the human poses efficiently. However, the accuracy and generality largely depend on the training data. Generative approaches usually formulate the problem as a tracking problem and adopt an explicit human model. Although arbitrary motions can be tracked, such systems usually have difficulties in adapting to different subjects and in dealing with tracking failures. In this thesis, an accurate, efficient and robust human pose tracking system from a single view camera is developed, mainly following a generative approach. A novel discriminative feature is also proposed and integrated into the tracking framework to improve the tracking performance. The human pose tracking system is proposed within a particle filtering framework. A reconfigurable skeleton model is constructed based on the Acclaim Skeleton File convention. A basic particle filter is first implemented for upper body tracking, which fuses time efficient cues from monocular sequences and achieves real-time tracking for constrained motions. Next, a 3D surface model is added to the skeleton model, and a full body tracking system is developed for more general and complex motions, assuming a stereo camera input. Partitioned sampling is adopted to deal with the high dimensionality problem, and the system is capable of running in near real-time. Multiple visual cues are investigated and compared, including a newly developed explicit depth cue. Based on the comparative analysis of cues, which reveals the importance of depth and good bottom-up features, a novel algorithm for detecting and identifying endpoint body parts from depth images is proposed. Inspired by the shape context concept, this thesis proposes a novel Local Shape Context (LSC) descriptor specifically for describing the shape features of body parts in depth images. This descriptor describes the local shape of different body parts with respect to a given reference point on a human silhouette, and is shown to be effective at detecting and classifying endpoint body parts. A new type of interest point is defined based on the LSC descriptor, and a hierarchical interest point selection algorithm is designed to further conserve computational resources. The detected endpoint body parts are then classified according to learned models based on the LSC feature. The algorithm is tested using a public dataset and achieves good accuracy with a 100Hz processing speed on a standard PC. Finally, the LSC descriptor is improved to be more generalized. Both the endpoint body parts and the limbs are detected simultaneously. The generalized algorithm is integrated into the tracking framework, which provides a very strong cue and enables tracking failure recovery. The skeleton model is also simplified to further increase the system efficiency. To evaluate the system on arbitrary motions quantitatively, a new dataset is designed and collected using a synchronized Kinect sensor and a marker based motion capture system, including 22 different motions from 5 human subjects. The system is capable of tracking full body motions accurately using a simple skeleton-only model in near real-time on a laptop PC before optimization.

This thesis develops new algorithms to produce 3D content from a single camera. Today, amateurs can use hand-held camcorders to capture and display the 3D world in 2D, using mature technologies. However, there is always a strong desire to record and re-explore the 3D world in 3D. To achieve this goal, current approaches usually make use of a camera array, which suffers from tedious setup and calibration processes, as well as lack of portability, limiting its application to lab experiments. In this thesis, I try to produce the 3D contents using a single camera, making it as simple as shooting pictures. It requires a new front end capturing device rather than a regular camcorder, as well as more sophisticated algorithms. First, in order to capture the highly detailed object surfaces, I designed and developed a depth camera based on a novel technique called light fall-off stereo (LFS). The LFS depth camera outputs color+depth image sequences and achieves 30 fps, which is necessary for capturing dynamic scenes. Based on the output color+depth images, I developed a new approach that builds 3D models of dynamic and deformable objects. While the camera can only capture part of a whole object at any instance, partial surfaces are assembled together to form a complete 3D model by a novel warping algorithm. Inspired by the success of single view 3D modeling, I extended my exploration into 2D-3D video conversion that does not utilize a depth camera. I developed a semi-automatic system that converts monocular videos into stereoscopic videos, via view synthesis. It combines motion analysis with user interaction, aiming to transfer as much depth inferring work from the user to the computer. I developed two new methods that analyze the optical flow in order to provide additional qualitative depth constraints. The automatically extracted depth information is presented in the user interface to assist with user labeling work. In this thesis, I developed new algorithms to produce 3D contents from a single camera. Depending on the input data, my algorithm can build high fidelity 3D models for dynamic and deformable objects if depth maps are provided. Otherwise, it can turn the video clips into stereoscopic video.

Recovery of human pose from videos has become a highly active research area in the last decade because of many attractive potential applications, such as surveillance, non-intrusive motion analysis and natural human machine interaction. Video based full body pose estimation is a very challenging task, because of the high degree of articulation of the human body, the large variety of possible human motions, and the diversity of human appearances. Methods for tackling this problem can be roughly categorized as either discriminative or generative. Discriminative methods can work on single images, and are able to recover the human poses efficiently. However, the accuracy and generality largely depend on the training data. Generative approaches usually formulate the problem as a tracking problem and adopt an explicit human model. Although arbitrary motions can be tracked, such systems usually have difficulties in adapting to different subjects and in dealing with tracking failures. In this thesis, an accurate, efficient and robust human pose tracking system from a single view camera is developed, mainly following a generative approach. A novel discriminative feature is also proposed and integrated into the tracking framework to improve the tracking performance. The human pose tracking system is proposed within a particle filtering framework. A reconfigurable skeleton model is constructed based on the Acclaim Skeleton File convention. A basic particle filter is first implemented for upper body tracking, which fuses time efficient cues from monocular sequences and achieves real-time tracking for constrained motions. Next, a 3D surface model is added to the skeleton model, and a full body tracking system is developed for more general and complex motions, assuming a stereo camera input. Partitioned sampling is adopted to deal with the high dimensionality problem, and the system is capable of running in near real-time. Multiple visual cues are investigated and compared, including a newly developed explicit depth cue. Based on the comparative analysis of cues, which reveals the importance of depth and good bottom-up features, a novel algorithm for detecting and identifying endpoint body parts from depth images is proposed. Inspired by the shape context concept, this thesis proposes a novel Local Shape Context (LSC) descriptor specifically for describing the shape features of body parts in depth images. This descriptor describes the local shape of different body parts with respect to a given reference point on a human silhouette, and is shown to be effective at detecting and classifying endpoint body parts. A new type of interest point is defined based on the LSC descriptor, and a hierarchical interest point selection algorithm is designed to further conserve computational resources. The detected endpoint body parts are then classified according to learned models based on the LSC feature. The algorithm is tested using a public dataset and achieves good accuracy with a 100Hz processing speed on a standard PC. Finally, the LSC descriptor is improved to be more generalized. Both the endpoint body parts and the limbs are detected simultaneously. The generalized algorithm is integrated into the tracking framework, which provides a very strong cue and enables tracking failure recovery. The skeleton model is also simplified to further increase the system efficiency. To evaluate the system on arbitrary motions quantitatively, a new dataset is designed and collected using a synchronized Kinect sensor and a marker based motion capture system, including 22 different motions from 5 human subjects. The system is capable of tracking full body motions accurately using a simple skeleton-only model in near real-time on a laptop PC before optimization.

In this thesis, we investigate the formulation, optimization and ambiguities in single-image 3D surface reconstruction from geometric and photometric constraints. We examine linear, discrete and quadratic constraints for shape from planar curves, shape from texture, and shape from shading. The problem of recovering 3D shape from the projection of planar curves on a surface is strongly motivated by perception studies. Applications include single-view modeling and uncalibrated structured light. When the curves intersect, the problem leads to a linear system for which a direct least-squares method is sensitive to noise. We derive a more stable solution and show examples where the same method produces plausible surfaces from the projection of parallel (non-intersecting) planar cross sections. The problem of reconstructing a smooth surface under constraints that have discrete ambiguities arise in areas such as shape from texture, shape from shading, photometric stereo and shape from defocus. While the problem is computationally hard, heuristics based on semidefinite programming may reveal the shape of the surface. Finally, we examine the shape from shading problem without boundary conditions as a polynomial system. This formulation allows, in generic cases, a complete solution for ideal polyhedral objects. For the general case we propose a semidefinite programming relaxation procedure, and an exact line search iterative procedure with a new smoothness term that favors folds at edges. We use this numerical technique to inspect shading ambiguities.

In this thesis, we investigate the formulation, optimization and ambiguities in single-image 3D surface reconstruction from geometric and photometric constraints. We examine linear, discrete and quadratic constraints for shape from planar curves, shape from texture, and shape from shading. The problem of recovering 3D shape from the projection of planar curves on a surface is strongly motivated by perception studies. Applications include single-view modeling and uncalibrated structured light. When the curves intersect, the problem leads to a linear system for which a direct least-squares method is sensitive to noise. We derive a more stable solution and show examples where the same method produces plausible surfaces from the projection of parallel (non-intersecting) planar cross sections. The problem of reconstructing a smooth surface under constraints that have discrete ambiguities arise in areas such as shape from texture, shape from shading, photometric stereo and shape from defocus. While the problem is computationally hard, heuristics based on semidefinite programming may reveal the shape of the surface. Finally, we examine the shape from shading problem without boundary conditions as a polynomial system. This formulation allows, in generic cases, a complete solution for ideal polyhedral objects. For the general case we propose a semidefinite programming relaxation procedure, and an exact line search iterative procedure with a new smoothness term that favors folds at edges. We use this numerical technique to inspect shading ambiguities.

ABSTRACT In our modern society, the design and implementation of intelligent equipments for autonomous physical services become more and more important. In line with this, the proposed Intelligent Vision Agent System, IVAS, is able to automatically detect and identify a target for a specific task by surveying human activities space. One of IVAS’ applications can be the adjustment of sink height for different height of people. Usually, the sink is fixed in one place, however, the height of sink could be too high to use for children. It becomes a real problem and may bring much of inconvenience and insecurity for the little boys and girls. The equipment for dynamical adjustment of sink height is rare in today’s society. The most common sink height adjustments are of two types. The first way is to place two sinks, one for adult and one for a kid individually. The second way is to use a spring device to adjust the height of sink. But, the both solutions have some limitations. The disadvantage of first method is that it takes too much space, and not all bathrooms can accommodate two sinks at the same time. The weakness of second way is a need to manually adjust the height of sink. In order to achieve an optimal design of adjustable sink high, the author uses a camera fixed on wall connected to an intelligent agent controlling suitable actuators. The camera captures a photo of the person who comes to the sink. The height of person can be estimated from the image. Furthermore, one makes use of this height value to find the suitable sink height for the user. Finally, the sink is descending or ascending by the lifting columns to adjust it to the different height of people. In this thesis, the author has implemented a method, which estimates the height of person from a single image. This technique is based on the single view metrology. Keywords: Sink, Height Measurement, Vision, Single View Metrology