3D Pose Research Articles

Investigating the efficacy of AI-enhanced telerehabilitation in sarcopenic older individuals.

This study explores the effectiveness of 3D pose estimation technology in Yi Jin Jing (a traditional Chinese exercise) interventions for sarcopenic older individuals. A randomized controlled trial involving 93 participants (mean age: 71.64 ± 7.09 years; 41 males and 52 females) divided into three groups: a face-to-face offline traditional training group (OFFG), a general remote online training group (ONG), and an AI-based online remote training group (AIONG). Participants in each group underwent their respective training programs. The effectiveness of the interventions was measured using Appendicular Skeletal Muscle Mass Index, Grip Strength, 6-meter Walking Speed, Timed-Up-and-Go Test, and Quality of Life assessments. Significant improvements were observed across all groups in ASMI, Grip Strength, 6-meter Walking Speed, TUGT, and QoL. However, there were no statistically significant differences between the groups in terms of the magnitude of these improvements. AIONG showed outcomes comparable to OFFG and ONG methods. AI-based telerehabilitation with 3D pose estimation is a viable and effective alternative for remote exercise interventions. It offers precise guidance and enhances the quality of rehabilitation training, demonstrating outcomes comparable to traditional and general online methods.

Fragment and torsion biasing algorithms for construction of small organic molecules in proteins using DOCK.

The computational construction of small organic molecules (de novo design), directly in a protein binding site, is an effective means for generating novel ligands tailored to fit the pocket environment. In this work, we present two new methods, which aim to improve de novo design outcomes using (1) biasing algorithms to prioritize selection and/or acceptance of fragments and torsions during growth, and (2) parallel-based clustering and pruning algorithms to remove duplicate molecules as candidate fragment are added. Large-scale testing encompassing thousands of simulations were employed to interrogate the methods in terms of multiple metrics which include numbers of duplicate molecules generated, pairwise-similarity, focused library reconstruction rates, fragment and torsion frequencies, fragment and torsion rank scores, interaction energy and drug-likeness scores, and 3D pose comparisons. The biasing algorithms, particularly those that include fragment and torsion components simultaneously, led to molecules that more closely mimicked the distributions of fragments and torsions found in drug-like libraries. The new parallel-based clustering and pruning algorithms, compared with the existing serial approach, also led to larger ensembles comprised of topologically unique molecules with much greater efficiency by removing redundant growth paths.

3D skeleton aware driver behavior recognition framework for autonomous driving system

The recognition of the driver’s behaviors inside an autonomous vehicle can effectively address emergency handling in autonomous driving and is crucial for ensuring the driver’s safety. Driver behavior recognition is a challenging task due to factors such as variations, diversities, complexities, and strong interferences in behaviors. In this paper, to realize the application in the autonomous driving scenes, a novel 3D skeleton aware behavior recognition framework is proposed to recognize various driver behaviors in autonomous driving systems. First, a 3D human pose estimation network (Pose-GTFNet) with temporal Transformer and spatial graph convolutional network (GCN) is designed to infer 3D human poses from 2D pose sequences. Second, based on the obtained 3D human pose sequences, a behavior recognition network (Beh-MSFNet) with multi-skeleton feature fusion is designed to recognize driver behaviors. In the experiments, the Pose-GTFNet and Beh-MSFNet can get the best performance compared with most state-of-the-art (SOTA) methods on the Human3.6M human pose dataset, JHMDB and SHREC action recognition dataset, respectively. In addition, the proposed driver behavior recognition framework can achieve SOTA performance on the Drive&Act and Driver-Skeleton driver behavior datasets.

Human Pose Classification Based on Pose Long Short-Term Memory

Objective: At present, pose classification research is a hot topic in the field of computer vision, playing a crucial role in surveillance security, and motion data mining. The research serves as a crucial follow-up to three dimensional (3D) pose estimation and makes a significant contribution to the advancement of artificial intelligence. Pose classification is a complex problem involving a large amount of complicated data, making its digital modeling and further processing challenging. Methods: This article proposes the Pose long short - term memory (LSTM) method to classify 3D poses reconstructed from 2D image sequences. Result: Compared with traditional methods, the experimental tests show that the performance of the Pose LSTM is more superior in all aspects. Conclusion: Pose LSTM is particularly effective for recognizing continuous action poses because they can capture temporal dependencies in sequential data. In the context of pose recognition, it can analyze the sequence of poses over time and understand how these poses transition from one to another, making it possible to accurately classify or predict actions that unfold over a series of frames.

DenseFusion-DA2: End-to-End Pose-Estimation Network Based on RGB-D Sensors and Multi-Channel Attention Mechanisms.

Notably, 6D pose estimation is a critical technology that enables robotics to perceive and interact with their operational environment. However, occlusion causes a loss of local features, which, in turn, restricts the estimation accuracy. To address these challenges, this paper proposes an end-to-end pose-estimation network based on a multi-channel attention mechanism, DA2Net. Firstly, a multi-channel attention mechanism, designated as "DA2Net", was devised using A2-Nets as its foundation. This mechanism is constructed in two steps. In the first step, the essential characteristics are extracted from the global feature space through the second-order attention pool. In the second step, a feature map is generated by the integration of position and channel attention. Subsequently, the extracted key features are assigned to each position of the feature map, enhancing both the feature representation capacity and the overall performance. Secondly, the designed attention mechanism is introduced into both the feature fusion and pose iterative refinement networks to enhance the network's capacity to acquire local features thus improving its overall performance. The experimental results demonstrated that the estimation accuracy of DenseFusion-DA2 on the LineMOD dataset was approximately 3.4% higher than that of DenseFusion. Furthermore, the estimation accuracy surpassed that of PoseCNN, PVNet, SSD6D, and PointFusion by 8.3%, 11.1%, 20.3%, and 23.8%, respectively. The estimation accuracy also shows a significant advantage on the Occluded LineMOD and HR-Vision datasets. This research not only presents a more efficient solution for robot perception but also introduces novel ideas and methods for technological advancements and applications in related fields.

3D Head Pose Estimation via Normal Maps: A Generalized Solution for Depth Image, Point Cloud, and Mesh

Head pose estimation plays a crucial role in various applications, including human–machine interaction, autonomous driving systems, and 3D reconstruction. Current methods address the problem primarily from a 2D perspective, which limits the efficient utilization of 3D information. Herein, a novel approach, called pose orientation‐aware network (POANet), which leverages normal maps for orientation information embedding, providing abundant and robust head pose information, is introduced. POANet incorporates the axial signal perception module and the rotation matrix perception module, these lightweight modules make the approach achieve state‐of‐the‐art (SOTA) performance with few computational costs. This method can directly analyze various topological 3D data without extensive preprocessing. For depth images, POANet outperforms existing methods on the Biwi Kinect head pose dataset, reducing the mean absolute error (MAE) by ≈30% compared to the SOTA methods. POANet is the first method to perform rigid head registration in a landmark‐free manner. It also incorporates few‐shot learning capabilities and achieves an MAE of about on the Headspace dataset. These features make POANet a superior alternative to traditional generalized Procrustes analysis for mesh data processing, offering enhanced convenience for human phenotype studies.

Using 3D Hand Pose Data in Recognizing Human–Object Interaction and User Identification for Extended Reality Systems

Object detection and action/gesture recognition have become imperative in security and surveillance fields, finding extensive applications in everyday life. Advancement in such technologies will help in furthering cybersecurity and extended reality systems through the accurate identification of users and their interactions, which plays a pivotal role in the security management of an entity and providing an immersive experience. Essentially, it enables the identification of human–object interaction to track actions and behaviors along with user identification. Yet, it is performed by traditional camera-based methods with high difficulties and challenges since occlusion, different camera viewpoints, and background noise lead to significant appearance variation. Deep learning techniques also demand large and labeled datasets and a large amount of computational power. In this paper, a novel approach to the recognition of human–object interactions and the identification of interacting users is proposed, based on three-dimensional hand pose data from an egocentric camera view. A multistage approach that integrates object detection with interaction recognition and user identification using the data from hand joints and vertices is proposed. Our approach uses a statistical attribute-based model for feature extraction and representation. The proposed technique is tested on the HOI4D dataset using the XGBoost classifier, achieving an average F1-score of 81% for human–object interaction and an average F1-score of 80% for user identification, hence proving to be effective. This technique is mostly targeted for extended reality systems, as proper interaction recognition and users identification are the keys to keeping systems secure and personalized. Its relevance extends into cybersecurity, augmented reality, virtual reality, and human–robot interactions, offering a potent solution for security enhancement along with enhancing interactivity in such systems.

3D human pose estimation with multi-hypotheses gated transformer

3D human pose estimation with multi-hypotheses gated transformer

DM-HAP: Diffusion model for accurate hand pose prediction

Forecasting hand poses is a challenging task due to inherent uncertainties, occlusions, and inaccuracies in 3D pose estimation. Diffusion models provide a promising direction for predicting precise 3D hand poses under noisy conditions. In this work, we introduce Dual-diffusion, an innovative framework for the precise prediction of future hand poses. Our approach leverages the strengths of diffusion models by framing hand pose forecasting as a reverse diffusion process, effectively addressing the complexities of noisy hand joint movements and their subtleties. To enhance the learning of hand pose representations, we propose a unique neural architecture that simultaneously captures both local and global features. This is achieved through the deployment of Global and Local Diffusion (GLD) blocks within our network, which facilitate the exchange of information between local and global features. This diffusion-based interaction enables the integration of global hand actions and local finger actions, leading to a more powerful representation learning approach. We evaluate the effectiveness of our Dual-diffusion method on three public 3D hand pose estimation datasets (MSRA, F-PHAB, and BigHand2.2M), and our approach outperforms previous methods.

A multi-stage approach for desired part grasping under complex backgrounds in human-robot collaborative assembly

Human-robot collaborative assembly can leverage the unique capabilities of humans and robots to provide a flexible and efficient way for complex tasks. In this context, robots are expected to be endowed with the perception ability to collaborate with humans. For flexible processes, robotic grasping for a desired assembly part from variable multi-parts is essential but remains challenging, especially in situations of complex backgrounds including disordered placement, occlusion, similarity, and large-scale differences in size. To improve the perception and decision-making abilities of robots in collaborative assembly, firstly, a multi-stage approach integrating 2D part recognition and 6D pose estimation with perspective-focusing is proposed for desired part grasping under complex backgrounds. Secondly, to efficiently and accurately recognize a desired part from variable multi-parts under complex backgrounds, based on RGB data, an enhanced detector is adopted, which lays the foundation for the subsequent pose estimation. Thirdly, based on point cloud data, a pose estimation method based on perspective-focusing is proposed to enhance the success rate and efficiency of pose acquisition. Additionally, a modular autonomous robotic grasping system is designed. Finally, taking decelerator assembly as a case, the proposed approaches are comprehensively validated and compared on a real collaborative robot platform. Under complex backgrounds with occlusion, our approach shows stable performance with an average grasping success rate of 90.0 % and an average cycle time of 6.4175 s. Our work is expected to improve robots’ perception and decision-making capabilities while further reducing the cycle time of collaborative assembly tasks.

On-Line Measurement of Tracking Poses of Heliostats in Concentrated Solar Power Plants.

The tracking pose of heliostats directly affects the stability and working efficiency of concentrated solar power (CSP) plants. Due to occlusion, over-exposure, and uneven illumination caused by mirror reflection, traditional image processing algorithms showed poor performances on the detection and segmentation of heliostats, which impede vision-based 3D measurement of tracking poses of heliostats. To tackle this issue, object detection using deep learning neural networks are exploited. An improved neural network based on YOLO-v5 framework has been designed to solve the on-line detection problem of heliostats. The model achieves a recognition accuracy of 99.7% for the test set, outperforming traditional methods significantly. Based on segmented results, the corner points of each heliostat are found out using Hough Transform and line intersection methods. The 3D poses of each heliostat are then solved out based on the image coordinates of specific feature points and the camera model. Experimental and field test results demonstrate the feasibility of this hybrid approach, which provides a low-cost solution for the monitoring and measurement of tracking poses of the heliostats in CSP.

Full-Body Pose Estimation of Humanoid Robots Using Head-Worn Cameras for Digital Human-Augmented Robotic Telepresence

We envision a telepresence system that enhances remote work by facilitating both physical and immersive visual interactions between individuals. However, during robot teleoperation, communication often lacks realism, as users see the robot’s body rather than the remote individual. To address this, we propose a method for overlaying a digital human model onto a humanoid robot using XR visualization, enabling an immersive 3D telepresence experience. Our approach employs a learning-based method to estimate the 2D poses of the humanoid robot from head-worn stereo views, leveraging a newly collected dataset of full-body poses for humanoid robots. The stereo 2D poses and sparse inertial measurements from the remote operator are optimized to compute 3D poses over time. The digital human is localized from the perspective of a continuously moving observer, utilizing the estimated 3D pose of the humanoid robot. Our moving camera-based pose estimation method does not rely on any markers or external knowledge of the robot’s status, effectively overcoming challenges such as marker occlusion, calibration issues, and dependencies on headset tracking errors. We demonstrate the system in a remote physical training scenario, achieving real-time performance at 40 fps, which enables simultaneous immersive and physical interactions. Experimental results show that our learning-based 3D pose estimation method, which operates without prior knowledge of the robot, significantly outperforms alternative approaches requiring the robot’s global pose, particularly during rapid headset movements, achieving markerless digital human augmentation from head-worn views.

Monocular 3D Multi-Person Pose Estimation for On-Site Joint Flexion Assessment: A Case of Extreme Knee Flexion Detection.

Work-related musculoskeletal disorders (WMSDs) represent a significant health challenge for workers in construction environments, often arising from prolonged exposure to ergonomic risks associated with manual labor, awkward postures, and repetitive motions. These conditions not only lead to diminished worker productivity but also incur substantial economic costs for employers and healthcare systems alike. Thus, there is an urgent need for effective tools to assess and mitigate these ergonomic risks. This study proposes a novel monocular 3D multi-person pose estimation method designed to enhance ergonomic risk assessments in construction environments. Leveraging advanced computer vision and deep learning techniques, this approach accurately captures and analyzes the spatial dynamics of workers' postures, with a focus on detecting extreme knee flexion, a critical indicator of work-related musculoskeletal disorders (WMSDs). A pilot study conducted on an actual construction site demonstrated the method's feasibility and effectiveness, achieving an accurate detection rate for extreme flexion incidents that closely aligned with supervisory observations and worker self-reports. The proposed monocular approach enables universal applicability and enhances ergonomic analysis through 3D pose estimation and group pose recognition for timely interventions. Future efforts will focus on improving robustness and integration with health monitoring to reduce WMSDs and promote worker health.

Robotic grasping method with 6D pose estimation and point cloud fusion

Robotic grasping method with 6D pose estimation and point cloud fusion

Enhancing Inter-AUV Perception: Adaptive 6-DOF Pose Estimation with Synthetic Images for AUV Swarm Sensing

The capabilities of AUV mutual perception and localization are crucial for the development of AUV swarm systems. We propose the AUV6D model, a synthetic image-based approach to enhance inter-AUV perception through 6D pose estimation. Due to the challenge of acquiring accurate 6D pose data, a dataset of simulated underwater images with precise pose labels was generated using Unity3D. Mask-CycleGAN technology was introduced to transform these simulated images into realistic synthetic images, addressing the scarcity of available underwater data. Furthermore, the Color Intermediate Domain Mapping strategy is proposed to ensure alignment across different image styles at pixel and feature levels, enhancing the adaptability of the pose estimation model. Additionally, the Salient Keypoint Vector Voting Mechanism was developed to improve the accuracy and robustness of underwater pose estimation, enabling precise localization even in the presence of occlusions. The experimental results demonstrated that our AUV6D model achieved millimeter-level localization precision and pose estimation errors within five degrees, showing exceptional performance in complex underwater environments. Navigation experiments with two AUVs further verified the model’s reliability for mutual 6D pose estimation. This research provides substantial technical support for more complex and precise collaborative operations for AUV swarms in the future.

Action-conditioned contrastive learning for 3D human pose and shape estimation in videos

The aim of this research is to estimate 3D human pose and shape in videos, which is a challenging task due to the complex nature of the human body and the wide range of possible pose and shape variations. This problem also poses difficulty in finding a satisfactory solution due to the trade-off between the accuracy and temporal consistency of the estimated 3D pose and shape. Thus previous researches have prioritized one objective over the other. In contrast, we propose a novel approach called the action-conditioned mesh recovery (ACMR) model, which improves accuracy without compromising temporal consistency by leveraging human action information. Our ACMR model outperforms existing methods that prioritize temporal consistency in terms of accuracy, while also achieving comparable temporal consistency with other state-of-the-art methods. Significantly, the action-conditioned learning process occurs only during training, requiring no additional resources at inference time, thereby enhancing performance without increasing computational demands.

Exploiting multi-transformer encoder with multiple-hypothesis aggregation via diffusion model for 3D human pose estimation

Exploiting multi-transformer encoder with multiple-hypothesis aggregation via diffusion model for 3D human pose estimation

Refined Prior Guided Category-Level 6D Pose Estimation and Its Application on Robotic Grasping

Estimating the 6D pose and size of objects is crucial in the task of visual grasping for robotic arms. Most current algorithms still require the 3D CAD model of the target object to match with the detected points, and they are unable to predict the object’s size, which significantly limits the generalizability of these methods. In this paper, we introduce category priors and extract high-dimensional abstract features from both the observed point cloud and the prior to predict the deformation matrix of the reconstructed point cloud and the dense correspondence between the reconstructed and observed point clouds. Furthermore, we propose a staged geometric correction and dense correspondence refinement mechanism to enhance the accuracy of regression. In addition, a novel lightweight attention module is introduced to further integrate the extracted features and identify potential correlations between the observed point cloud and the category prior. Ultimately, the object’s translation, rotation, and size are obtained by mapping the reconstructed point cloud to a normalized canonical coordinate system. Through extensive experiments, we demonstrate that our algorithm outperforms existing methods in terms of performance and accuracy on commonly used benchmarks for this type of problem. Additionally, we implement the algorithm in robotic arm-grasping simulations, further validating its effectiveness.

Six-Degree-of-Freedom Pose Estimation Method for Multi-Source Feature Points Based on Fully Convolutional Neural Network

An object’s six-degree-of-freedom (6DoF) pose information has great importance in various fields. Existing methods of pose estimation usually detect two-dimensional (2D)-three-dimensional (3D) feature point pairs, and directly estimates the pose information through Perspective-n-Point (PnP) algorithms. However, this approach ignores the spatial association between pixels, making it difficult to obtain high-precision results. In order to apply pose estimation based on deep learning methods to real-world scenarios, we hope to design a method that is robust enough in more complex scenarios. Therefore, we introduce a method for 3D object pose estimation from color images based on farthest point sampling (FPS) and object 3D bounding box. This method detects the 2D projection of 3D feature points through a convolutional neural network, matches it with the 3D model of the object, and then uses the PnP algorithm to restore the feature point pair to the object pose. Due to the global nature of the bounding box, this approach can be considered effective even in partially occluded or complex environments. In addition, we propose a heatmap suppression method based on weighted coordinates to further improve the prediction accuracy of feature points and the accuracy of the PnP algorithm in solving the pose position. Compared with other algorithms, this method has higher accuracy and better robustness. Our method yielded 93.8% of the ADD(-s) metrics on the Linemod dataset and 47.7% of the ADD(-s) metrics on the Occlusion Linemod dataset. These results show that our method is more effective than existing methods in pose estimation of large objects.

DBMHT: A double-branch multi-hypothesis transformer for 3D human pose estimation in video

The estimation of 3D human poses from monocular videos presents a significant challenge. The existing methods face the problems of deep ambiguity and self-occlusion. To overcome these problems, we propose a Double-Branch Multi-Hypothesis Transformer (DBMHT). In detail, we utilize a Double-Branch architecture to capture temporal and spatial information and generate multiple hypotheses. To merge these hypotheses, we adopt a lightweight module to integrate spatial and temporal representations. The DBMHT can not only capture spatial information from each joint in the human body and temporal information from each frame in the video but also merge multiple hypotheses that have different spatio-temporal information. Comprehensive evaluation on two challenging datasets (i.e. Human3.6M and MPI-INF-3DHP) demonstrates the superior performance of DBMHT, marking it as a robust and efficient approach for accurate 3D HPE in dynamic scenarios. The results show that our model surpasses the state-of-the-art approach by 1.9% MPJPE with ground truth 2D keypoints as input.