Automatic video analysis and motion estimation for physical activity classification (original) (raw)
Related papers
Physical activity recognition based on motion in images acquired by a wearable camera
Neurocomputing, 2011
A new technique to extract and evaluate physical activity patterns from image sequences captured by a wearable camera is presented in this paper. Unlike standard activity recognition schemes, the video data captured by our device do not include the wearer him/herself. The physical activity of the wearer, such as walking or exercising, is analyzed indirectly through the camera motion extracted from the acquired video frames. Two key tasks, pixel correspondence identification and motion feature extraction, are studied to recognize activity patterns. We utilize a multiscale approach to identify pixel correspondences. When compared with the existing methods such as the Good Features detector and the Speed-up Robust Feature (SURF) detector, our technique is more accurate and computationally efficient. Once the pixel correspondences are determined which define representative motion vectors, we build a set of activity pattern features based on motion statistics in each frame. Finally, the physical activity of the person wearing a camera is determined according to the global motion distribution in the video. Our algorithms are tested using different machine learning techniques such as the K-Nearest Neighbor (KNN), Naive Bayesian and Support Vector Machine (SVM). The results show that many types of physical activities can be recognized from field acquired real-world video. Our results also indicate that, with a design of specific motion features in the input vectors, different classifiers can be used successfully with similar performances.
Recognizing physical activity from ego-motion of a camera
Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2010
A new image based activity recognition method for a person wearing a video camera below the neck is presented in this paper. The wearable device is used to capture video data in front of the wearer. Although the wearer never appears in the video, his or her physical activity is analyzed and recognized using the recorded scene changes resulting from the motion of the wearer. Correspondence features are extracted from adjacent frames and inaccurate matches are removed based on a set of constraints imposed by the camera model. Motion histograms are defined and calculated within a frame and we define a new feature called accumulated motion distribution derived from motion statistics in each frame. A Support Vector Machine (SVM) classifier is trained with this feature and used to classify physical activities in different scenes. Our results show that different types of activities can be recognized in low resolution, field acquired real-world video.
Differentiating two daily activities through analysis of short ambulatory video clips
MeMeA 2013 - IEEE International Symposium on Medical Measurements and Applications, Proceedings, 2013
Automatically detecting daily activities using wearable smartphones would provide valuable information to clinicians. While accelerometer data is effective in this area, classifying stair ascent can be difficult. In this paper, video content analysis is performed on short videos captured from a wearable smartphone in order to distinguish between level ground walking and stair climbing. High contrast image features, such as corners, were tracked across consecutive video frames to create feature paths. Computing the median of the slope of the paths in each frame revealed substantial differences, in both magnitude and variation over time, for stair climbing as opposed to walking. A time series of median slope values was produced for each video clip, and the number of local maxima and minima above a threshold of 1.0 were computed. Results revealed that the number of peaks during stair climbing were substantially larger than walking and, therefore, could be used as a feature for distinguishing between these two activities.
Toward Automatic Activity Classification and Movement Assessment During a Sports Training Session
IEEE Internet of Things Journal, 2015
Motion analysis technologies have been widely used to monitor the potential for injury and enhance athlete performance. However, most of these technologies are expensive, can only be used in laboratory environments and examine only a few trials of each movement action. In this paper, we present a novel ambulatory motion analysis framework using wearable inertial sensors to accurately assess all of an athlete's activities in real training environment. We firstly present a system that automatically classifies a large range of training activities using the Discrete Wavelet Transform (DWT) in conjunction with a Random forest classifier. The classifier is capable of successfully classifying various activities with up to 98% accuracy. Secondly, a computationally efficient gradient descent algorithm is used to estimate the relative orientations of the wearable inertial sensors mounted on the shank, thigh and pelvis of a subject, from which the flexion-extension knee and hip angles are calculated. These angles, along with sacrum impact accelerations, are automatically extracted for each stride during jogging. Finally, normative data is generated and used to determine if a subject's movement technique differed to the normative data in order to identify potential injury related factors. For the joint angle data this is achieved using a curve-shift registration technique. It is envisaged that the proposed framework could be utilized for accurate and automatic sports activity classification and reliable movement technique evaluation in various unconstrained environments for both injury management and performance enhancement.
Recognizing Physical Activities using Wearable Devices
IGI Global eBooks, 2017
Physical activity is a major part of the user's context for wearable computing applications. The System should be able to acquire the user's physical activities using body worn sensors. The authors propose developing a personal activity recognition system that is practical, reliable, and can be used for healthcare related applications. They propose to use the wearable device which is a readymade, light weight, small and easy to use device for identifying physical activities (i.e. lying, sitting, walking, standing, cycling, running, ascending stairs and descending stairs), fitness studio activities (i.e. using elliptical trainer, butterfly, bench-press and pull down) and swimming techniques (i.e., dolphin, backstroke , breast-stroke and free-style) using machine learning algorithms. In this chapter, the authors present an approach to build a system that exhibits this property and provides evidence based on user studies. Their results indicate that the system has a good accuracy rate.
International Journal of Scientific Research in Science, Engineering and Technology, 2022
The human activity monitoring system helps to differentiate a person's physical actions such as walking, clapping, shaking hands etc. Activity awareness is the foundation for the development of many potential applications for health, wellness and sports. HAR has a variety of uses because of its impact on health. Helps users improve quality of life in areas such as aged care, daily logging, personal fitness software. Personal Performance Recognition is a field for identifying basic human activity and is currently being used in various fields where important information about an individual's ability to work and lifestyle. As the famous saying goes "Exercise not only changes our body it changes our mind, our mood, and our attitude". Fitness is a practice today. Everyone wants to be fit, to be beautiful, and to be healthy. But during this epidemic, not everyone can hire a coach or go to the gym. Another option is wearable devices that not everyone can afford. This paper proposed an AI trainer model. The proposed model used by anyone regardless of age and health status. The AI model uses Personal Status Evaluation. It is a popular method and determines the location and posture of the human body. This technique creates important points in the human body and is based on the fact that it creates a virtual skeletal structure in the 2D dimension. Featured is a live video taken from a person's webcam and the output captures location marks or key points in the human body. The AI trainer specifies the calculation and timing of the settings that a person must perform. It also specifies errors and feedback if any. This paper provides a way to use the stop rate that works on the CPU to get the correct points. Based on points touch and other curls (biceps) are calculated. This paper proposes a method that uses OpenCV to use a stand-alone model.
Physical Activity Recognition Based on Machine Learning
Proceedings of the 29th Minisymposium, 2022
The following paper presents a comparison study of various machine learning techniques in recognition of activities of daily living (ADL), with special attention being given to movements during human falling and the distinction among various types of falls. The motivation for the development of physical activity recognition algorithm includes keeping track of users' activities in real-time, and possible diagnostics of unwanted and unexpected movements and/or events. The activities recorded and processed in this study include various types of daily activities, such as walking, running, etc., while fall activities include falling forward, falling backward, falling left and right (front fall, back fall and side fall). The algorithm was trained on two publicly available datasets containing signals from an accelerometer, a magnetometer and a gyroscope.
7th Vienna International Conference on Mathematical Modelling, 2012
Sensor and computing technologies provide people with information on their performance and load when doing sports. In order to automatically give advices on how to continue exercising and/or to adjust the sports equipment during the physical activity, intelligent devices are required. These devices rely on models for recognition and classification of patterns in the motion currently performed. Different methods and models, such as Neural Networks, Hidden Markov models or Support Vector Machines have proven to be applicable for this purpose. Pros and cons of the different approaches are discussed. Practical applications are presented and experiences reported.
2007 11th IEEE International Symposium on Wearable Computers, 2007
In this paper, we present a real-time algorithm for automatic recognition of not only physical activities, but also, in some cases, their intensities, using five triaxial wireless accelerometers and a wireless heart rate monitor. The algorithm has been evaluated using datasets consisting of 30 physical gymnasium activities collected from a total of 21 people at two different labs. On these activities, we have obtained a recognition accuracy performance of 94.6% using subject-dependent training and 56.3% using subjectindependent training. The addition of heart rate data improves subject-dependent recognition accuracy only by 1.2% and subject-independent recognition only by 2.1%. When recognizing activity type without differentiating intensity levels, we obtain a subjectindependent performance of 80.6%. We discuss why heart rate data has such little discriminatory power.