Search for: All records

ObjectiveThis study aims to improve workers’ postures and thus reduce the risk of musculoskeletal disorders in human-robot collaboration by developing a novel model-free reinforcement learning method. BackgroundHuman-robot collaboration has been a flourishing work configuration in recent years. Yet, it could lead to work-related musculoskeletal disorders if the collaborative tasks result in awkward postures for workers. MethodsThe proposed approach follows two steps: first, a 3D human skeleton reconstruction method was adopted to calculate workers’ continuous awkward posture (CAP) score; second, an online gradient-based reinforcement learning algorithm was designed to dynamically improve workers’ CAP score by adjusting the positions and orientations of the robot end effector. ResultsIn an empirical experiment, the proposed approach can significantly improve the CAP scores of the participants during a human-robot collaboration task when compared with the scenarios where robot and participants worked together at a fixed position or at the individual elbow height. The questionnaire outcomes also showed that the working posture resulted from the proposed approach was preferred by the participants. ConclusionThe proposed model-free reinforcement learning method can learn the optimal worker postures without the need for specific biomechanical models. The data-driven nature of this method can make it adaptive to provide personalized optimal work posture. ApplicationThe proposed method can be applied to improve the occupational safety in robot-implemented factories. Specifically, the personalized robot working positions and orientations can proactively reduce exposure to awkward postures that increase the risk of musculoskeletal disorders. The algorithm can also reactively protect workers by reducing the workload in specific joints. 

Advances in robotics have contributed to the prevalence of human-robot collaboration (HRC). Working and interacting with collaborative robots in close proximity can be psychologically stressful. Therefore, it is important to understand the impacts of human-robot interaction (HRI) on mental stress to promote psychological well-being at the workplace. To this end, this study investigated how the HRI presence, complexity, and modality affect psychological stress in humans and discussed possible HRI design criteria during HRC. An experimental setup was implemented in which human operators worked with a collaborative robot on a Lego assembly task, using different interaction paradigms involving pressing buttons, showing hand gestures, and giving verbal commands. The NASA-Task Load Index, as a subjective measure, and the physiological galvanic skin conductance response, as an objective measure, were used to assess the levels of mental stress. The results revealed that the introduction of interactions during HRC helped reduce mental stress and that complex interactions resulted in higher mental stress than simple interactions. Meanwhile, the use of certain interaction modalities, such as verbal commands or hand gestures, led to significantly higher mental stress than pressing buttons, while no significant difference on mental stress was found between showing hand gestures and giving verbal commands. 

Decision-based black-box attacks often necessitate a large number of queries to craft an adversarial example. Moreover, decision-based attacks based on querying boundary points in the estimated normal vector direction often suffer from inefficiency and convergence issues. In this paper, we propose a novel query-efficient \b curvature-aware \b geometric decision-based \b black-box \b attack (CGBA) that conducts boundary search along a semicircular path on a restricted 2D plane to ensure finding a boundary point successfully irrespective of the boundary curvature. While the proposed CGBA attack can work effectively for an arbitrary decision boundary, it is particularly efficient in exploiting the low curvature to craft high-quality adversarial examples, which is widely seen and experimentally verified in commonly used classifiers under non-targeted attacks. In contrast, the decision boundaries often exhibit higher curvature under targeted attacks. Thus, we develop a new query-efficient variant, CGBA-H, that is adapted for the targeted attack. In addition, we further design an algorithm to obtain a better initial boundary point at the expense of some extra queries, which considerably enhances the performance of the targeted attack. Extensive experiments are conducted to evaluate the performance of our proposed methods against some well-known classifiers on the ImageNet and CIFAR10 datasets, demonstrating the superiority of CGBA and CGBA-H over state-of-the-art non-targeted and targeted attacks, respectively. 

The main challenge of monocular 3D object detection is the accurate localization of 3D center. Motivated by a new and strong observation that this challenge can be remedied by a 3D-space local-grid search scheme in an ideal case, we propose a stage-wise approach, which combines the information flow from 2D-to-3D (3D bounding box proposal generation with a single 2D image) and 3D-to-2D (proposal verification by denoising with 3D-to-2D contexts) in a topdown manner. Specifically, we first obtain initial proposals from off-the-shelf backbone monocular 3D detectors. Then, we generate a 3D anchor space by local-grid sampling from the initial proposals. Finally, we perform 3D bounding box denoising at the 3D-to-2D proposal verification stage. To effectively learn discriminative features for denoising highly overlapped proposals, this paper presents a method of using the Perceiver I/O model [20] to fuse the 3D-to-2D geometric information and the 2D appearance information. With the encoded latent representation of a proposal, the verification head is implemented with a self-attention module. Our method, named as MonoXiver, is generic and can be easily adapted to any backbone monocular 3D detectors. Experimental results on the well-established KITTI dataset and the challenging large-scale Waymo dataset show that MonoXiver consistently achieves improvement with limited computation overhead. 

In recent years, there has been a trend to adopt human-robot collaboration (HRC) in the industry. In previous studies, computer vision-aided human pose reconstruction is applied to find the optimal position of point of operation in HRC that can reduce workers’ musculoskeletal disorder (MSD) risks due to awkward working postures. However, the reconstruction of human pose through computer-vision may fail due to the complexity of the workplace environment. In this study, we propose a data-driven method for optimizing the position of point of operation during HRC. A conditional variational auto-encoder (cVAE) model-based approach is adopted, which includes three steps. First, a cVAE model was trained using an open-access multimodal human posture dataset. After training, this model can output a simulated worker posture of which the hand position can reach a given position of point of operation. Next, an awkward posture score is calculated to evaluate MSD risks associated with the generated postures with a variety of positions of point of operation. The position of point of operation that is associated with a minimum awkward posture score is then selected for an HRC task. An experiment was conducted to validate the effectiveness of this method. According to the findings, the proposed method produced a point of operation position that was similar to the one chosen by participants through subjective selection, with an average difference of 4.5 cm. 

Vision Transformers (ViTs) are built on the assumption of treating image patches as “visual tokens” and learn patch-to-patch attention. The patch embedding based tokenizer has a semantic gap with respect to its counterpart, the textual tokenizer. The patch-to-patch attention suffers from the quadratic complexity issue, and also makes it non-trivial to explain learned ViTs. To address these issues in ViT, this paper proposes to learn Patch-to-Cluster attention (PaCa) in ViT. Queries in our PaCa-ViT starts with patches, while keys and values are directly based on clustering (with a predefined small number of clusters). The clusters are learned end-to-end, leading to better tokenizers and inducing joint clustering-for-attention and attention-for-clustering for better and interpretable models. The quadratic complexity is relaxed to linear complexity. The proposed PaCa module is used in designing efficient and interpretable ViT backbones and semantic segmentation head networks. In experiments, the proposed methods are tested on ImageNet-1k image classification, MS-COCO object detection and instance segmentation and MIT-ADE20k semantic segmentation. Compared with the prior art, it obtains better performance in all the three benchmarks than the SWin [32] and the PVTs [47], [48] by significant margins in ImageNet-1k and MIT-ADE20k. It is also significantly more efficient than PVT models in MS-COCO and MIT-ADE20k due to the linear complexity. The learned clusters are semantically meaningful. Code and model checkpoints are available at https:/github.com/iVMCL/PaCaViT. 

This paper presents a neural incremental Structure-from-Motion (SfM) approach, Level-S2fM, which estimates the camera poses and scene geometry from a set of uncalibrated images by learning coordinate MLPs for the implicit surfaces and the radiance fields from the established key-point correspondences. Our novel formulation poses some new challenges due to inevitable two-view and few-view configurations in the incremental SfM pipeline, which complicates the optimization of coordinate MLPs for volumetric neural rendering with unknown camera poses. Nevertheless, we demonstrate that the strong inductive basis conveying in the 2D correspondences is promising to tackle those challenges by exploiting the relationship between the ray sampling schemes. Based on this, we revisit the pipeline of incremental SfM and renew the key components, including two-view geometry initialization, the camera poses registration, the 3D points triangulation, and Bundle Adjustment, with a fresh perspective based on neural implicit surfaces. By unifying the scene geometry in small MLP networks through coordinate MLPs, our Level-S2fM treats the zero-level set of the implicit surface as an informative top-down regularization to manage the reconstructed 3D points, reject the outliers in correspondences via querying SDF, and refine the estimated geometries by NBA (Neural BA). Not only does our Level-S2fM lead to promising results on camera pose estimation and scene geometry reconstruction, but it also shows a promising way for neural implicit rendering without knowing camera extrinsic beforehand. 

This article presents Holistically-Attracted Wireframe Parsing (HAWP), a method for geometric analysis of 2D images containing wireframes formed by line segments and junctions. HAWP utilizes a parsimonious Holistic Attraction (HAT) field representation that encodes line segments using a closed-form 4D geometric vector field. The proposed HAWP consists of three sequential components empowered by end-to-end and HAT-driven designs: (1) generating a dense set of line segments from HAT fields and endpoint proposals from heatmaps, (2) binding the dense line segments to sparse endpoint proposals to produce initial wireframes, and (3) filtering false positive proposals through a novel endpoint-decoupled line-of-interest aligning (EPD LOIAlign) module that captures the co-occurrence between endpoint proposals and HAT fields for better verification. Thanks to our novel designs, HAWPv2 shows strong performance in fully supervised learning, while HAWPv3 excels in self-supervised learning, achieving superior repeatability scores and efficient training (24 GPU hours on a single GPU). Furthermore, HAWPv3 exhibits a promising potential for wireframe parsing in out-of-distribution images without providing ground truth labels of wireframes. 

This paper studies the challenging two-view 3D reconstruction problem in a rigorous sparse-view configuration, which is suffering from insufficient correspondences in the input image pairs for camera pose estimation. We present a novel Neural One-PlanE RANSAC framework (termed NOPE-SAC in short) that exerts excellent capability of neural networks to learn one-plane pose hypotheses from 3D plane correspondences. Building on the top of a Siamese network for plane detection, our NOPE-SAC first generates putative plane correspondences with a coarse initial pose. It then feeds the learned 3D plane correspondences into shared MLPs to estimate the one-plane camera pose hypotheses, which are subsequently reweighed in a RANSAC manner to obtain the final camera pose. Because the neural one-plane pose minimizes the number of plane correspondences for adaptive pose hypotheses generation, it enables stable pose voting and reliable pose refinement with a few of plane correspondences for the sparse-view inputs. In the experiments, we demonstrate that our NOPE-SAC significantly improves the camera pose estimation for the two-view inputs with severe viewpoint changes, setting several new state-of-the-art performances on two challenging benchmarks, i.e., MatterPort3D and ScanNet, for sparse-view 3D reconstruction. The source code is released at https://github.com/IceTTTb/NopeSAC for reproducible research.