Visual experience recognition using adaptive support vector machine

Video has more information than the isolated images. Processing, analyzing and understanding of contents present in videos are becoming very important. Consumer videos are generally captured by amateurs using handheld cameras of events and it contains considerable camera motion, occlusion, cluttered background, and large intraclass variations within the same type of events, making their visual cues highly variable and less discriminant. So visual event recognition is an extremely challenging task in computer vision. A visual event recognition framework for consumer videos is framed by leveraging a large amount of loosely labeled web videos. The videos are divided into training and testing sets manually. A simple method called the Aligned Space-Time Pyramid Matching method was proposed to effectively measure the distances between two video clips from different domains. Each video is divided into space-time volumes over multiple levels. A new transfer learning method is referred to as Adaptive Multiple Kernel Learning fuse the information from multiple pyramid levels, features, and copes with the considerable variation in feature distributions between videos from two domains web video domain and consumer video domain.With the help of MATLAB Simulink videos are divided and compared with web domain videos. The inputs are taken from the Kodak data set and the results are given in the form of MATLAB simulation.

Study and application of computer information big data in basketball vision system using high-definition camera motion data capture

Contemporary computer big data technology is developing rapidly, and it has played a role in promoting the intelligent development of sports. Based on this research background, the thesis revolves around the application of computer big data artificial intelligence in the field of basketball training. The study found that computerized big data in basketball training and teaching is mainly reflected in the following aspects: high-definition camera motion data capture, player analysis to obtain assisted training system, etc. The final paper uses a basketball vision system as a case study to analyze specific applications of big data for computer information.

Multiview Running and Walking Gait Analysis using the First and Third Person Data

Gait recognition, which allows to recognise someone from a distance, has a lot of applications. The gait of a person is commonly used as a biometric approach to identify or categorise them by gender and age. Biometric systems are a fast-growing subject that necessitates the development of creative solutions to problems that have plagued previous attempts. By initially looking at the design of a gait detection system, two different types of gait datasets have been reported and presented in this paper. While running at the same time, the first person (FP) data containing the camera motion gathered from the movement of the volunteer’s body and the third person (TP) data captured from a distant view were recorded. The dataset contains a total of 23 participants (14 males and 9 females). The discussion is expanded to include a comparison of the results obtained using TP and FP data, as well as an examination of the physiological motions recorded while running and walking.

Biomechanical Effects of Flamenco Footwork

Footwork is one of the basic features of flamenco dancing and is performed in traditional high-heeled shoes. The purpose of this study was to analyse the mechanical profile of flamenco dancing in terms of vertical ground reaction force, and knee joint kinematics of the supporting limb in footwork technique in order to understand causes which predispose injuries derived from the practice of flamenco dancing. The participant in our study was a professional female flamenco dancer (34 years, 58 kg, 1.65 m) who performed the ZAP 3 test, a sequence of single strikes of the feet performed continuously for 15 s. 3D lower extremity kinematic data were collected using a five-camera motion analysis system (Vicon; Oxford Metrics Ltd., Oxford, UK). Ground reaction forces were recorded using a Kistler force plate. Our analysis was based on 30 cycles of each lower limb consisting of 177 footwork steps. The vertical component of the ground reaction force did not reveal any significant differences between the left and the right limb. The most dynamic strike was provided by the heel (twice the participant's body weight). The mean angular displacement of the supporting limb’s knee was ~27°. Results reveal that these impacts could make the knee joint more prone to injuries.

Kinematics of the head and associated vertebral artery length changes during high-velocity, low-amplitude cervical spine manipulation

Background Cervical spine manipulation (CSM) is a frequently used treatment for neck pain. Despite its demonstrated efficacy, concerns regarding CSM safety remain. The purpose of this study was to quantify the angular displacements of the head relative to the sternum and the associated vertebral artery (VA) length changes during the thrust phase of CSM. Methods Bilateral rotation and lateral flexion CSM procedures were delivered from C1 to C7 to three male cadaveric donors. For each CSM the force-time profile was recorded using a thin, flexible pressure pad (100-200Hz), to determine the timing of the thrust. Three dimensional displacements of the head relative to the sternum were recorded using an eight-camera motion analysis system (120-240Hz) and angular displacements of the head relative to the sternum were computed in Matlab. Positive kinematic values indicate flexion, left lateral flexion, and left rotation. Ipsilateral refers to the same side as the clinician's contact and contralateral, the opposite. Length changes of the VA were recorded using eight piezoelectric ultrasound crystals, inserted along the entire vessel. VA length changes were calculated as D=(L1-L0)/L0, where L0= length of the whole VA (sum of segmental lengths) or the V3 segment at CSM thrust onset; L1= whole VA or V3 length at peak force during the CSM thrust. Results VA length changes during the thrust phase were greatest with ipsilateral rotation CSM (producing contralateral head rotation): [mean ± SD (range)] whole artery [1.3 ± 1.0 (-0.4 to 3.3%)]; and V3 segment [2.6 ± 3.6 (-0.4 to 11.6%)]. For ipsilateral rotation CSM, head angular displacements relative to the sternum during the thrust were: flexion/extension [1.2 ± 3.4 (-6.6 to 7.6º)]; rotation [-10.2 ± 3.5 (-16.1 to -3.7º)]; and lateral flexion [8.8 ± 3.0 (2.5 to 14.1º)]. Conclusion Mean head angular displacements and VA length changes were small during CSM thrusts. Of the four different CSM measured, mean VA length changes were largest during rotation procedures. This suggests that if clinicians wish to limit VA length changes, consideration should be given to the type of CSM used.

Estimation of Walking Speed and Its Spatiotemporal Determinants Using a Single Inertial Sensor Worn on the Thigh: From Healthy to Hemiparetic Walking

We present the use of a single inertial measurement unit (IMU) worn on the thigh to produce stride-by-stride estimates of walking speed and its spatiotemporal determinants (i.e., stride time and stride length). Ten healthy and eight post-stroke individuals completed a 6-min walk test with an 18-camera motion capture system used for ground truth measurements. Subject-specific estimation models were trained to estimate walking speed using the polar radius extracted from phase portraits produced from the IMU-measured thigh angular position and velocity. Consecutive flexion peaks in the thigh angular position data were used to define each stride and compute stride times. Stride-by-stride estimates of walking speed and stride time were then used to compute stride length. In both the healthy and post-stroke cohorts, low error and high consistency were observed for the IMU estimates of walking speed (MAE < 0.035 m/s; ICC > 0.98), stride time (MAE < 30 ms; ICC > 0.97), and stride length (MAE < 0.037 m; ICC > 0.96). This study advances the use of a single wearable sensor to accurately estimate walking speed and its spatiotemporal determinants during both healthy and hemiparetic walking.

Effects of Shoe Midsole Hardness on Lower Extremity Biomechanics during Jump Rope in Healthy Males

This study investigated differences in lower extremity muscle activations and vertical stiffness during a 2.2 Hz jump rope exercise with different midsole hardnesses (45, 50, 55, and 60 Shores C). Twelve healthy male participants wore customized shoes with different hardness midsoles and performed jump rope exercises in a random order. A nine-camera motion analysis system (150 Hz), a force platform (1500 Hz), and a wireless electromyography (EMG) system (Noraxon, 1500 Hz) were used to measure the biomechanical parameters during the jump rope exercise. The biceps femoris %MVC of barefoot participants was significantly greater than that of those wearing the 45 Shores C (p = 0.048) and 55 Shores C (p = 0.009) midsole 100 ms before landing. The vastus medialis %MVC of barefoot participants was significantly greater than that of those wearing the 55 C midsole (p = 0.005). Nonsignificant differences in vertical stiffness were found between midsole hardnesses and barefoot. Lower extremity muscle activation differed between conditions. The results of this study indicate that for repetitive activities that entail multiple impacts, sports shoes with a low midsole hardness (e.g., 50 Shores C or 45 Shores C) may be appropriate. It is important to provide customers with information regarding midsole hardness in shoe product labeling so that they properly consider the function of the shoes.