ERNet: An Efficient and Reliable Human-Object Interaction Detection Network.

Human-Object Interaction (HOI) detection recognizes how persons interact with objects, which is advantageous in autonomous systems such as self-driving vehicles and collaborative robots. However, current HOI detectors are often plagued by model inefficiency and unreliability when making a prediction, which consequently limits its potential for real-world scenarios. In this paper, we address these challenges by proposing ERNet, an end-to-end trainable convolutional-transformer network for HOI detection. The proposed model employs an efficient multi-scale deformable attention to effectively capture vital HOI features. We also put forward a novel detection attention module to adaptively generate semantically rich instance and interaction tokens. These tokens undergo pre-emptive detections to produce initial region and vector proposals that also serve as queries which enhances the feature refinement process in the transformer decoders. Several impactful enhancements are also applied to improve the HOI representation learning. Additionally, we utilize a predictive uncertainty estimation framework in the instance and interaction classification heads to quantify the uncertainty behind each prediction. By doing so, we can accurately and reliably predict HOIs even under challenging scenarios. Experiment results on the HICO-Det, V-COCO, and HOI-A datasets demonstrate that the proposed model achieves state-of-the-art performance in detection accuracy and training efficiency. Codes are publicly available at https://github.com/Monash-CyPhi-AI-Research-Lab/ernet.

Vitrimer-Like Shape Memory Polymers: Characterization and Applications in Reshaping and Manufacturing.

The shape memory effect (SME) refers to the ability of a material to recover its original shape, but only in the presence of a right stimulus. Most polymers, either thermo-plastic or thermoset, can have the SME, although the actual shape memory performance varies according to the exact material and how the material is processed. Vitrimer, which is between thermoset and thermo-plastic, is featured by the reversible cross-linking. Vitrimer-like shape memory polymers (SMPs) combine the vitrimer-like behavior (associated with dissociative covalent adaptable networks) and SME, and can be utilized to achieve many novel functions that are difficult to be realized by conventional polymers. In the first part of this paper, a commercial polymer is used to demonstrate how to characterize the vitrimer-like behavior based on the heating-responsive SME. In the second part, a series of cases are presented to reveal the potential applications of vitrimer-like SMPs and their composites. It is concluded that the vitrimer-like feature not only enables many new ways in reshaping polymers, but also can bring forward new approaches in manufacturing, such as, rapid 3D printing in solid state on space/air/sea missions.