RESUMO
In overcoming the challenges faced in adapting to paired real-world data, recent unsupervised single image deraining (SID) methods have proven capable of accomplishing notably acceptable deraining performance. However, the previous methods usually fail to produce a high quality rain-free image due to neglecting sufficient attention to semantic representation and the image content, which results in the inability to completely separate the content from the rain layer. In this paper, we develop a novel cycle contrastive adversarial framework for unsupervised SID, which mainly consists of cycle contrastive learning (CCL) and location contrastive learning (LCL). Specifically, CCL achieves high-quality image reconstruction and rain-layer stripping by pulling similar features together while pushing dissimilar features further in both semantic and discriminant latent spaces. Meanwhile, LCL implicitly constrains the mutual information of the same location of different exemplars to maintain the content information. In addition, recently inspired by the powerful Segment Anything Model (SAM) that can effectively extract widely applicable semantic structural details, we formulate a structural-consistency regularization to fine-tune our network using SAM. Apart from this, we attempt to introduce vision transformer (VIT) into our network architecture to further improve the performance. In our designed transformer-based GAN, to obtain a stronger representation, we propose a multi-layer channel compression attention module (MCCAM) to extract a richer feature. Equipped with the above techniques, our proposed unsupervised SID algorithm, called CCLformer, can show advantageous image deraining performance. Extensive experiments demonstrate both the superiority of our method and the effectiveness of each module in CCLformer. The code is available at https://github.com/zhihefang/CCLGAN.
Assuntos
Processamento de Imagem Assistida por Computador , Redes Neurais de Computação , Processamento de Imagem Assistida por Computador/métodos , Algoritmos , Aprendizado de Máquina não Supervisionado , Humanos , SemânticaRESUMO
Three-dimensional (3D) bioprinting has advantages for constructing artificial skin tissues in replicating the structures and functions of native skin. Although many studies have presented improved effect of printing skin substitutes in wound healing, using hydrogel inks to fabricate 3D bioprinting architectures with complicated structures, mimicking mechanical properties, and appropriate cellular environments is still challenging. Inspired by collagen nanofibers withstanding stress and regulating cell behavior, a patterned nanofibrous film was introduced to the printed hydrogel scaffold to fabricate a composite artificial skin substitute (CASS). The artificial dermis was printed using gelatin-hyaluronan hybrid hydrogels containing human dermal fibroblasts with gradient porosity and integrated with patterned nanofibrous films simultaneously, while the artificial epidermis was formed by seeding human keratinocytes upon the dermis. The collagen-mimicking nanofibrous film effectively improved the tensile strength and fracture resistance of the CASS, making it sewable for firm implantation into skin defects. Meanwhile, the patterned nanofibrous film also provided the biological cues to guide cell behavior. Consequently, CASS could effectively accelerate the regeneration of large-area skin defects in mouse and pig models by promoting re-epithelialization and collagen deposition. This research developed an effective strategy to prepare composite bioprinting architectures for enhancing mechanical property and regulating cell behavior, and CASS could be a promising skin substitute for treating large-area skin defects.
Assuntos
Bioimpressão , Nanofibras , Impressão Tridimensional , Pele Artificial , Humanos , Nanofibras/química , Animais , Camundongos , Suínos , Hidrogéis/química , Fibroblastos/citologia , Engenharia Tecidual , Queratinócitos/citologia , Alicerces Teciduais/química , Ácido Hialurônico/química , Gelatina/químicaRESUMO
The growth of data and Internet of Things challenges traditional hardware, which encounters efficiency and power issues owing to separate functional units for sensors, memory, and computation. In this study, we designed an α-phase indium selenide (α-In2Se3) transistor, which is a two-dimensional ferroelectric semiconductor as the channel material, to create artificial optic-neural and electro-neural synapses, enabling cutting-edge processing-in-sensor (PIS) and computing-in-memory (CIM) functionalities. As an optic-neural synapse for low-level sensory processing, the α-In2Se3 transistor exhibits a high photoresponsivity (2855 A/W) and detectivity (2.91 × 1014 Jones), facilitating efficient feature extraction. For high-level processing tasks as an electro-neural synapse, it offers a fast program/erase speed of 40 ns/50 µs and ultralow energy consumption of 0.37 aJ/spike. An AI vision system using α-In2Se3 transistors has been demonstrated. It achieved an impressive recognition accuracy of 92.63% within 12 epochs owing to the synergistic combination of the PIS and CIM functionalities. This study demonstrates the potential of the α-In2Se3 transistor in future vision hardware, enhancing processing, power efficiency, and AI applications.