Search

Over the past few years, a growing number of researchers have dedicated their efforts to focusing on temporal modeling. The advent of transformer-based methods has notably advanced the field of 2D image-based vision tasks. However, with respect to 3D video tasks such as action recognition, applying temporal transformations directly to video data significantly increases both computational and memory demands. This surge in resource consumption is due to the multiplication of data patches and the added complexity of self-aware computations. Accordingly, building efficient and precise 3D self-attentive models for video content represents as a major challenge for transformers. In our research, we introduce an Long and Short-term Temporal Difference Vision Transformer (LS-VIT). This method incorporates short-term motion details into images by weighting the difference across several consecutive frames, thereby equipping the original image with the ability to model short-term motions. Concurrently, we integrate a module designed to understand long-term motion details. This module enhances the model's capacity for long-term motion modeling by directly integrating temporal differences from various segments via motion excitation. Our thorough analysis confirms that the LS-VIT achieves high recognition accuracy across multiple benchmarks (e.g., UCF101, HMDB51, Kinetics-400). These research results indicate that LS-VIT has the potential for further optimization, which can improve real-time performance and action prediction capabilities.

Prion–like domains (PrLDs), a unique type of low–complexity domain (LCD) or intrinsically disordered region (IDR), have been shown to mediate protein liquid–liquid phase separation (LLPS). Recent research has increasingly focused on how prion–like proteins (PrLPs) regulate plant growth, development, and stress responses. This review provides a comprehensive overview of plant PrLPs. We analyze the structural features of PrLPs and the mechanisms by which PrLPs undergo LLPS. Through gene ontology (GO) analysis, we highlight the diverse molecular functions of PrLPs and explore how PrLPs influence plant development and stress responses via phase separation. Finally, we address unresolved questions about PrLP regulatory mechanisms, offering prospects for future research.

Abstract With the increasing demand for multispectral information acquisition, infrared multispectral imaging technology that is inexpensive and can be miniaturized and integrated into other devices has received extensive attention. However, the widespread usage of such photodetectors is still limited by the high cost of epitaxial semiconductors and complex cryogenic cooling systems. Here, we demonstrate a noncooled two-color infrared photodetector that can provide temporal-spatial coexisting spectral blackbody detection at both near-infrared and mid-infrared wavelengths. This photodetector consists of vertically stacked back-to-back diode structures. The two-color signals can be effectively separated to achieve ultralow crosstalk of ~0.05% by controlling the built-in electric field depending on the intermediate layer, which acts as an electron-collecting layer and hole-blocking barrier. The impressive performance of the two-color photodetector is verified by the specific detectivity (D*) of 6.4 × 109 cm Hz1/2 W−1 at 3.5 μm and room temperature, as well as the promising NIR/MWIR two-color infrared imaging and absolute temperature detection.

Abstract Machine vision systems (MVSs) are an important component of intelligent systems, such as autonomous vehicles and robots. However, with the continuous increase in data and new application scenarios, new requirements are put forward for the next generation of MVS. There is an urgent need to find new material systems to complement the existing semiconductor technology based on thin‐film materials, and new architectures must be explored to improve efficiency. Because of their unique physical properties, two‐dimensional (2D) materials have received extensive attention for use in MVSs, especially in biomimetic ones: the human visual system, which can process complex visual information with low power consumption, provides a model for next‐generation MVSs. This review paper summarizes the progress and challenges of applying 2D material photodetectors in sense‐memory‐computational integration and biomimetic image sensors for machine vision.

Isolation technology has been successfully applied in seismic migration. With increasing of seismic demand, seismic performance of isolation structures subjected to very-rare earthquakes need further improvement. However, the isolation layer generally lacks sufficient deformation ability under very-rare earthquakes due to the deformation limit of classical isolation bearing. In order to circumvent the difficulty, this paper develops two new isolation bearings, namely super-large displacement rotation friction pendulum bearing (SLDRFPB) and super-large displacement translation friction pendulum bearing (SLDTFPB). By setting spherical shells with large span and large radius, large horizontal displacement and small horizontal stiffness can be achieved. Safety of the isolation layer and the isolation effect of the superstructure can be greatly improved. SLDTFPB differs from SLDRFPB in the motion state of the superstructure and space utilization of the isolation layer, thus SLDRFPB and SLDTFPB are suitable for structures with different requirements. Due to rotation of the superstructure with SLDRFPB or sliding frames in SLDTFPB, the traditional design method of friction pendulum bearing is no longer suitable. We present a new procedure to accurately and conveniently evaluate seismic performance of two developed bearings. Numerical simulation shows that the seismic response of both the superstructure and isolation layer is small. Developed SLDRFPB and SLDTFPB have sufficient emergency capacity and isolation resilience when subjected to very-rare earthquakes.