Measures of human movement dynamics can predict outcomes like injury risk or musculoskeletal disease progression. However, these measures are rarely quantified in clinical practice due to the prohibitive cost, time, and expertise required. Here we present and validate OpenCap, an open-source platform for computing movement dynamics using videos captured from smartphones. OpenCap’s web application enables users to collect synchronous videos and visualize movement data that is automatically processed in the cloud, thereby eliminating the need for specialized hardware, software, and expertise. We show that OpenCap accurately predicts dynamic measures, like muscle activations, joint loads, and joint moments, which can be used to screen for disease risk, evaluate intervention efficacy, assess between-group movement differences, and inform rehabilitation decisions. Additionally, we demonstrate OpenCap’s practical utility through a 100-subject field study, where a clinician using OpenCap estimated movement dynamics 25 times faster than a laboratory-based approach at less than 1% of the cost. By democratizing access to human movement analysis, OpenCap can accelerate the incorporation of biomechanical metrics into large-scale research studies, clinical trials, and clinical practice.

/pdf/opencap-3d-human-movement-dynamics-from-smartphone-videos-nb9m7blo.pdf

OpenCap: 3D human movement dynamics from smartphone videos

This paper proposes a simple baseline framework for video-based 2D/3D human pose estimation that can achieve 10 times efficiency improvement over existing works without any performance degradation, named DeciWatch. Unlike current solutions that estimate each frame in a video, DeciWatch introduces a simple yet effective sample-denoise-recover framework that only watches sparsely sampled frames, taking advantage of the continuity of human motions and the lightweight pose representation. Specifically, DeciWatch uniformly samples less than 10% video frames for detailed estimation, denoises the estimated 2D/3D poses with an efficient Transformer architecture, and then accurately recovers the rest of the frames using another Transformer-based network. Comprehensive experimental results on three video-based human pose estimation and body mesh recovery tasks with four datasets validate the efficiency and effectiveness of DeciWatch. Code is available at https://github.com/cure-lab/DeciWatch.

DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimation

The analysis of human gait is an important tool in medicine and rehabilitation to evaluate the effects and the progression of neurological diseases resulting in neuromotor disorders. In these fields, the gold standard techniques adopted to perform gait analysis rely on motion capture systems and markers. However, these systems present drawbacks: they are expensive, time consuming and they can affect the naturalness of the motion. For these reasons, in the last few years, considerable effort has been spent to study and implement markerless systems based on videography for gait analysis. Unfortunately, only few studies quantitatively compare the differences between markerless and marker-based systems in 3D settings. This work presented a new RGB video-based markerless system leveraging computer vision and deep learning to perform 3D gait analysis. These results were compared with those obtained by a marker-based motion capture system. To this end, we acquired simultaneously with the two systems a multimodal dataset of 16 people repeatedly walking in an indoor environment. With the two methods we obtained similar spatio-temporal parameters. The joint angles were comparable, except for a slight underestimation of the maximum flexion for ankle and knee angles. Taking together these results highlighted the possibility to adopt markerless technique for gait analysis.

/pdf/markerless-vs-marker-based-gait-analysis-a-proof-of-concept-3atroxcc.pdf

Markerless vs. Marker-Based Gait Analysis: A Proof of Concept Study

The very fast technological progress provokes the creation of a big volume of the information that can be fed in different forms. There are various science directions that use high dimensional data sets for the analysis. In this paper we presented the few aspects of the "knowledge discovery in databases" (KDD) process related to the Data Mining stage in astronomy, analyzed and reviewed Data Mining approaches. We presented the examples of astronomical sources of Big Data, instruments, information types, processing algorithms that can be used for the Data Mining process in astronomy. The paper deals with applying the CoLiTec (Collection Light Technology) software for the online processing of the different types of astronomical information using the Data Mining approach. This is achieved by using of the developed OnLine Data Analysis System (OLDAS), which helps with solving of the Data Mining tasks, like clustering, classification, and identification.

Data Mining of the Astronomical Images by the CoLiTec Software

Background Kinematic analysis of the upper limbs is a good way to assess and monitor recovery in individuals with stroke, but remains little used in clinical routine due to its low feasibility. The aim of this study is to assess the validity and reliability of the Kinect v2 for the assessment of 17 kinematic variables commonly used in the analysis of upper limb reaching in stroke. Methods 26 healthy participants performed seated hand-reaching tasks while holding a dumbbell to induce a behaviour similar to that of a person with a stroke. 3D upper limb and trunk motion were simultaneously recorded with the Kinect v2 (Microsoft, USA) and with the VICON (OxfordMetrics, UK), the latter being the reference system. For each kinematic outcome, the validity of the Kinect was assessed with ICC, linear regression and Bland & Altman plots. Results The Kinect assesses trunk compensations, hand range of motion, movement time and mean velocity with a moderate to excellent reliability. In contrast, elbow and shoulder range of motion, time to peak velocity and path length ratio have a poor to moderate reliability, indicating that these variables should be interpreted with caution. Finally, instantaneous hand and elbow tracking are not precise enough to reliably assess Cartesian and angular kinematics over time, rendering variables such as the number of velocity peaks and the peak hand velocity unusable. Conclusions Thanks to its ease of use and markerless properties, the Kinect can be used in clinical routine for semi-automated quantitative diagnostics guiding individualised rehabilitation of the upper limb. However, engineers and therapists must bear in mind the limitations of the Kinect for the instantaneous tracking of the hand and elbow.

/pdf/validity-and-reliability-of-kinect-v2-for-quantifying-upper-1bfuji5w.pdf

Validity and Reliability of Kinect v2 for Quantifying Upper Body Kinematics during Seated Reaching

Human pose estimation is a very active research field, stimulated by its important applications in robotics, entertainment or health and sports sciences, among others. Advances in convolutional networks triggered noticeable improvements in 2D pose estimation, leading modern 3D markerless motion capture techniques to an average error per joint of 20 mm. However, with the proliferation of methods, it is becoming increasingly difficult to make an informed choice. Here, we review the leading human pose estimation methods of the past five years, focusing on metrics, benchmarks and method structures. We propose a taxonomy based on accuracy, speed and robustness that we use to classify de methods and derive directions for future research.

A review of 3D human pose estimation algorithms for markerless motion capture

Ridges (resp., valley) are the useful geometric features due to their wide varieties of applications, mainly in image analysis problems such as object detection, image segmentation, scene understanding, etc. Many characterizations have contributed to formalize the ridge notion. The signification of each characterization rely however on its actual application. The objective analysis of ridge characterized as thin and complex image structure is thus essentially important, for choosing which parameter’s values correspond to the suitable configuration to obtain accurate results and optimal performance. In this article an extensive analysis followed by a supervised and objective comparison of different filtering-based ridge detection techniques is led. Furthermore, the optimal parameter configuration of each filtering techniques aimed for image salient feature analysis tool have been objectively investigated, where each chosen filter’s parameters corresponds to the width of the desired ridge or valley. At last, the comparative evaluations and analysis results are reported on both synthetic images, distorted with various types of noises and real images.

Ridge Detection by Image Filtering Techniques: A Review and an Objective Analysis

Among the common image structures, line feature is the extensively used geometric structure for various image processing applications, including the analysis of biomedical image with blood vessels highlighting, graph-shape structures, cracks detection, satellite images or remote sensing data. Multi-scale processing of line feature is essentially required for the extraction of more relevant information or line structures of heterogeneous widths. In this paper, a multi-scale filtering-based line detection approach using second-order semi-Gaussian anisotropic kernel is proposed. Meanwhile, a strategy is introduced to calculate the strength of the observed line feature across the different scales. The proposed technique is evaluated on real images by using their tied hand-labeled images. Finally, the experimental results and comparison of images containing different line feature widths with state-of-the-art techniques have sufficiently supported the effectiveness of our technique.

A Multi-scale Line Feature Detection Using Second Order Semi-Gaussian Filters

Ridges and valleys are the principle geometric features for their diverse applications, especially in image analysis problems such as segmentation, object detection, etc. Numerous characterizations have contributed to formalize the ridge and valley theory. The significance of each characterization rely however on its practical usefulness in a particular application. The objective comparison and evaluation of ridgeness/valleyness characterized as thin and complex image structure is thus crucially important, for choosing, which parameter’s values correspond to the optimal configuration to obtain accurate results and best performance. This paper presents a supervised and objective comparison of different filtering-based ridge detectors. Moreover, the optimal parameter configuration of each filtering techniques have been objectively investigated.

Philippe Montesinos

Papers

A review of 3D human pose estimation algorithms for markerless motion capture

A review of 3D human pose estimation algorithms for markerless motion capture

Ridge Detection by Image Filtering Techniques: A Review and an Objective Analysis

A Multi-scale Line Feature Detection Using Second Order Semi-Gaussian Filters

An Objective Comparison of Ridge/Valley Detectors by Image Filtering