COWO: towards real-time spatiotemporal action localization in videos

Purpose The purpose of this paper is to provide a fast and accurate network for spatiotemporal action localization in videos. It detects human actions both in time and space simultaneously in real-time, which is applicable in real-world scenarios such as safety monitoring and collaborative assembly. Design/methodology/approach This paper design an end-to-end deep learning network called collaborator only watch once (COWO). COWO recognizes the ongoing human activities in real-time with enhanced accuracy. COWO inherits from the architecture of you only watch once (YOWO), known to be the best performing network for online action localization to date, but with three major structural modifications: COWO enhances the intraclass compactness and enlarges the interclass separability in the feature level. A new correlation channel fusion and attention mechanism are designed based on the Pearson correlation coefficient. Accordingly, a correction loss function is designed. This function minimizes the same class distance and enhances the intraclass compactness. Use a probabilistic K-means clustering technique for selecting the initial seed points. The idea behind this is that the initial distance between cluster centers should be as considerable as possible. CIOU regression loss function is applied instead of the Smooth L1 loss function to help the model converge stably. Findings COWO outperforms the original YOWO with improvements of frame mAP 3% and 2.1% at a speed of 35.12 fps. Compared with the two-stream, T-CNN, C3D, the improvement is about 5% and 14.5% when applied to J-HMDB-21, UCF101-24 and AGOT data sets. Originality/value COWO extends more flexibility for assembly scenarios as it perceives spatiotemporal human actions in real-time. It contributes to many real-world scenarios such as safety monitoring and collaborative assembly.

An Approach Based on VR to Design Industrial Human-Robot Collaborative Workstations

This paper presents an integrated approach for the design of human–robot collaborative workstations in industrial shop floors. In particular, the paper presents how to use virtual reality (VR) technologies to support designers in the creation of interactive workstation prototypes and in early validation of design outcomes. VR allows designers to consider and evaluate in advance the overall user experience, adopting a user-centered perspective. The proposed approach relies on two levels: the first allows designers to have an automatic generation and organization of the workstation physical layout in VR, starting from a conceptual description of its functionalities and required tools; the second aims at supporting designers during the design of Human–Machine Interfaces (HMIs) by interaction mapping, HMI prototyping and testing in VR. The proposed approach has been applied on two realistic industrial case studies related to the design of an intensive warehouse and a collaborative assembly workstation for automotive industry, respectively. The two case studies demonstrate how the approach is suited for early prototyping of complex environments and human-machine interactions by taking into account the user experience from the early phases of design.

C-ALB (Collaborative Assembly Line Balancing): a new approach in cobot solutions

Collaborative robots can be a proper solution to improve the throughput of manual systems without reducing their flexibility. To effectively use cobots in productive systems, it is fundamental to develop a suitable task allocation model that considers collaboration. Hence, we present a model for collaborative assembly line balancing (C-ALB) which considers paralleling tasks and collaboration in the balancing resolution. Indexes that take into account both the product and process characteristics are defined to evaluate the quality of the proposed task allocation model and comparing it to others. The results confirm the influence of the product characteristics on the system performance, leading to the definition of a new paradigm for product design.