Search

Pose prediction is to predict future poses given a window of previous poses. In this paper, we propose a new problem that predicts poses using 3D positions of skeletal sequences.Different from the traditional pose prediction based on mocap frames, this problem is convenient to use in real applications due to its simple sensors to capture data. We also present a new framework, pseudo-image sequence evolution-based 3D pose prediction, to address this new problem. Specifically, a skeletal representation is proposed by transforming a 3D skeletal sequence into an image sequence, which can model different correlations among different joints. With this image-based skeletal representation, we model the pose prediction as the evolution of an image sequence. Moreover, a novel inference network is proposed to predict multiple future poses in a non-recursive manner using decoders with independent parameters. In contrast to the recursive sequence-to-sequence model, we can improve the computational efficiency and avoid error accumulations significantly. Extensive experiments are carried out on two benchmark datasets (e.g., G3D and FNTU). The proposed method achieves state-of-the-art performance on both datasets, which demonstrates the effectiveness of our proposed method. [ABSTRACT FROM AUTHOR]

The practical application of the lithium anode in lithium metal batteries (LMBs) has been hindered by the uncontrollable growth of lithium dendrite and the high volumetric change during cycling. Herein, the in situ formed three-dimensional (3D) lithium–boron (Li–B) alloy is suggested as an excellent alternative to the Li metal, in which the 3D LiB skeleton can mitigate the growth of Li dendrites and volumetric change. In this study, the Li–B alloy anodes with different B contents were manufactured by high-temperature melting. It was found that the boron content had a significant effect on the electrochemical performance of the Li–B alloy. The Li–B alloy with the least B content (10 wt%, 10LiB) demonstrated the lowest overpotential of 0.0852 V after 300 h and the lowest interface resistance. However, the full cell with 15LiB as the anode displayed the best cycling performance of 115 mAh·g−1 after 100 cycles with a columbic efficiency greater than 97%. The obtained results suggest that the in situ formed three-dimensional Li–B alloy anode can be an excellent alternative to the Li anode via tuning B contents for next-generation high energy density LMBs. [ABSTRACT FROM AUTHOR]

To communicate with people, robots and vision-based interactive systems often need to understand human activities in advance before the activity is performed completely. This early prediction of the activities will help them take proper near future steps to fulfill a realistic interactive session with humans. However, predicting activities in advance is a very challenging task, because some activities are simple while others are complex and comprised of several smaller atomic sub-activities. In this paper, we propose a method capable of early prediction of simple and complex human activities by formulating it as a structured prediction task using probabilistic graphical models (PGM). We use skeletons captured from low-cost depth sensors as high-level descriptions of the human body. Using 3D skeletons, our method will be robust to the environmental factors. Our proposed model is a fully observed PGM coupled with a clustering scheme to remove the dependency of our model to the number-of-middle-states hyperparameter. We test our method on three popular datasets: CAD-60, UT-Kinect, and Florence 3D and obtain accuracies of 97.6% , 100% and 96.11%, respectively. These datasets cover both simple and complex activities. When only half of the clip is observed, we achieve 93.33% and 96.9% accuracy on CAD-60 and UT-Kinect datasets, respectively. [ABSTRACT FROM AUTHOR]

Understanding 3D object structure from a single image is an important but challenging task in computer vision, mostly due to the lack of 3D object annotations to real images. Previous research tackled this problem by either searching for a 3D shape that best explains 2D annotations, or training purely on synthetic data with ground truth 3D information. In this work, we propose 3D INterpreter Networks (3D-INN), an end-to-end trainable framework that sequentially estimates 2D keypoint heatmaps and 3D object skeletons and poses. Our system learns from both 2D-annotated real images and synthetic 3D data. This is made possible mainly by two technical innovations. First, heatmaps of 2D keypoints serve as an intermediate representation to connect real and synthetic data. 3D-INN is trained on real images to estimate 2D keypoint heatmaps from an input image; it then predicts 3D object structure from heatmaps using knowledge learned from synthetic 3D shapes. By doing so, 3D-INN benefits from the variation and abundance of synthetic 3D objects, without suffering from the domain difference between real and synthesized images, often due to imperfect rendering. Second, we propose a Projection Layer, mapping estimated 3D structure back to 2D. During training, it ensures 3D-INN to predict 3D structure whose projection is consistent with the 2D annotations to real images. Experiments show that the proposed system performs well on both 2D keypoint estimation and 3D structure recovery. We also demonstrate that the recovered 3D information has wide vision applications, such as image retrieval. [ABSTRACT FROM AUTHOR]

In this paper, we present an image classification approach to action recognition with 3D skeleton videos. First, we propose a video domain translation-scale invariant image mapping, which transforms the 3D skeleton videos to color images, namely skeleton images. Second, a multi-scale dilated convolutional neural network (CNN) is designed for the classification of the skeleton images. Our multi-scale dilated CNN model could effectively improve the frequency adaptiveness and exploit the discriminative temporal-spatial cues for the skeleton images. Even though the skeleton images are very different from natural images, we show that the fine-tuning strategy still works well. Furthermore, we propose different kinds of data augmentation strategies to improve the generalization and robustness of our method. Experimental results on popular benchmark datasets such as NTU RGB + D, UTD-MHAD, MSRC-12 and G3D demonstrate the superiority of our approach, which outperforms the state-of-the-art methods by a large margin. [ABSTRACT FROM AUTHOR]

This paper presents a new 3D gait recognition method that utilizes the kinect skeleton data for representing the gait signature. We propose to use two new features, namely joint relative distance (JRD) and joint relative angle (JRA), which are robust against view and pose variations. The relevance of each JRD and JRA sequence in representing human gait is evaluated using a genetic algorithm. We also introduce a dynamic time warping-based kernel that takes a collection of JRD or JRA sequences as parameters and computes a dissimilarity measure between the training and the unknown sample. The proposed kernel can effectively handle variable walking speed without any need of extra pre-processing. In addition, we propose a rank-level fusion of JRD and JRA features that can boost the overall recognition performance greatly. The effectiveness of the proposed method is evaluated using a 3D skeletal gait database captured with a Kinect v2 sensor. In our experiments, rank level fusion of joint relative distance (JRD) and joint relative angle (JRA) achieves promising results, as compared against only JRD and only JRA-based gait recognition. [ABSTRACT FROM AUTHOR]

This paper deals with two techniques to represent relevant information from volumetric vascular images in a more compact format. The images are obtained with contrast-enhanced magnetic resonance angiography (MRA). After segmentation of the vessels, the curve skeleton is extracted by an algorithm based on the distance transformation. The algorithm first reduces the original object to a surface skeleton and then to a curve skeleton, after which “pruning” can be performed to remove irrelevant small branches. Applying this procedure to MRA data from the pelvic arteries resulted in a good description of the tree structure of the vessels with a much smaller number of voxels. To detect stenoses, 2D projections such as maximum intensity projection (MIP) are usually employed, but these often fail to demonstrate a stenosis if the projection angle is not suitably chosen. A new presentation method surrounds each voxel in the distance-labeled curve skeleton of the segmented vascular tree with a ball whose radius represents the minimum vessel radius at that level. Experiments with synthetic data indicate that stenoses invisible in an ordinary projection may be seen with this technique. It is concluded that the distance-labelled curve skeleton seems to be useful for visualizing variations in vessel calibre and in the future possibly also for quantification of arterial stenoses. [ABSTRACT FROM AUTHOR]

Four soil cores (length, 20 cm; diameter, 16 cm) were sampled in a Swiss pre-Alpine meadow with high earthworm abundance (>400 individuals/m2); two cores were taken in October 1993 and the other two cores in April 1994. The cores were described using computer assisted tomography which gives a series of section images every 3 mm. A method for reconstructing the three-dimensional (3D) skeleton of the earthworm burrow system is presented and discussed. This method provides an image of the structural organisation of the burrow system and was found to be adequately sensitive for use in ecological and functional studies. The seasonal variation of these 3D skeletons was investigated using two approaches, i.e. the analysis of: (1) global burrow system characteristics, and (2) individual burrow characteristics. At the scale of the global burrow system no difference was found between seasons (same number of burrows and same total burrow length) except for the vertical segment distribution, which was homogeneous in spring and decreased with depth in the fall. The study of individual burrow characteristics revealed that burrows tended to be more vertical in spring and that their branching intensity was higher in this season. [ABSTRACT FROM AUTHOR]