Results 71 to 80 of about 1,582,815 (295)

Multimodal Optoelectronic Synapses for Neuromorphic Intelligence: Coupling Light, Electrical, Thermal, and Beyond

Advanced Functional Materials, EarlyView.
This review revisits optoelectronic neuromorphic devices that extend from unimodal sensing toward multimodal perception. Five types of stimuli‐to‐output information flow that are classified based on the device roles within the system and applications are summarized and discussed.
Jihyun Shin, Junhyung Cho, Wangmyung Choi, Jaehyun Hur, Jae Kyeong Jeong, Ulrike Kraft, Hocheon Yoo   +6 more
wiley   +1 more source

Light‐Induced Field‐Tunneling Synapses in Solution‐Processed Van Der Waals Heterostructures for Scalable, Retina‐Inspired Optical Sensing

Advanced Functional Materials, EarlyView.
A scalable, solution‐processed WSe2/ZrO2‐x van der Waals heterostructure realizes a light‐induced field‐tunneling synapse (LIFTS) that activates exclusively under bright illumination, emulating the photopic adaptation of the human retina at the device level.
Kijeong Nam, Jihyun Kim, Soobeen Ji, In Soo Kim, Sejin Kim, Eungseon Yeon, Do Kyung Hwang, Iva Plutnarová, Zdeněk Sofer, Sang Min Won, Joohoon Kang   +10 more
wiley   +1 more source

Grafting Vision Transformers

2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
Vision Transformers (ViTs) have recently become the state-of-the-art across many computer vision tasks. In contrast to convolutional networks (CNNs), ViTs enable global information sharing even within shallow layers of a network, i.e., among high-resolution features. However, this perk was later overlooked with the success of pyramid architectures such
Jongwoo Park 0003, Kumara Kahatapitiya, Donghyun Kim 0006, Shivchander Sudalairaj, Quanfu Fan, Michael S. Ryoo   +5 more
openaire   +2 more sources

Towards Robust Vision Transformer

Recent advances on Vision Transformer (ViT) and its improved variants have shown that self-attention-based networks surpass traditional Convolutional Neural Networks (CNNs) in most vision tasks.
Mao, Xiaofeng, Xue, Hui, He, Yuan, Chen, Yuefeng, Li, Xiaodan, Ye, Shaokai, Duan, Ranjie, Qi, Gege   +7 more
core   +1 more source

Emerging Post‐CMOS Hardware Neurons for Brain‐Inspired Computing: Devices, Circuits, and System Integration

Advanced Functional Materials, EarlyView.
The physical realization of artificial neurons is a critical challenge for energy‐efficient neuromorphic computing. This review presents a comprehensive analysis of the evolution of artificial neuron implementations from conventional CMOS to emerging post‐CMOS technologies.
Kannan Udaya Mohanan, Hocheon Yoo, Benoît H. Lessard, Ioannis Kymissis, Chang‐Hyun Kim   +4 more
wiley   +1 more source

SE‐Swin: An improved Swin‐Transfomer network of self‐ensemble feature extraction framework for image retrieval

IET Image Processing
The Swin‐Transformer is a variant of the Vision Transformer, which constructs a hierarchical Transformer that computes representations with shifted windows and window multi‐head self‐attention.
Yixuan Xu, Xianbing Wang, Hua Zhang, Hai Lin   +3 more
doaj   +1 more source

Sub‐1 ms Optoelectronic Synapse Based on {ZnCdO/ZnO} Multilayer Structure for High‐Speed Neuromorphic Vision Systems

Advanced Functional Materials, EarlyView.
Eu‐doped {ZnCdO/ZnO} structures grown on Si are developed, showcasing dual‐functionality controlled by europium doping. While high europium concentration transforms the device into an ultrafast, self‐powered photodetector, low‐doped structures can be used as an optoelectronic synapse.
Igor Perlikowski, Eunika Zielony, Adrian Kaim, Ewa Przeździecka   +3 more
wiley   +1 more source

A Hierarchical ViT With Dynamic Window Shift Unit and Curriculum Learning for Remote Sensing Image Scene Classification

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
The remote sensing image (RSI) scene classification is currently a popular research topic among many remote sensing tasks. However, RSI scene classification still faces challenges such as complex multiscale key features concentrated in different local ...
Yi Liu, Xiang Wu, Jihuan Ren, Jiacun Wang, Yuming Bo, Yuanhao Wang   +5 more
doaj   +1 more source

A Sensorimotor Vision Transformer

CoRR
This paper presents the Sensorimotor Transformer (SMT), a vision model inspired by human saccadic eye movements that prioritize high-saliency regions in visual input to enhance computational efficiency and reduce memory consumption. Unlike traditional models that process all image patches uniformly, SMT identifies and selects the most salient patches ...
Konrad Gadzicki, Kerstin Schill, Christoph Zetzsche   +2 more
openaire   +3 more sources

The Effect of Hydrophilic Headgroup Chemistry on Peptoid‐Directed Mineralization of CaCO3

Advanced Functional Materials, EarlyView.
Amphiphilic peptoids alter the pathway and outcome of calcium carbonate crystallization, which progresses through amorphous calcium carbonate (ACC) to vaterite to calcite rhombs in the absence of peptoids. In contrast, amine‐terminated peptoids produce vaterite disks and carboxyl‐terminated peptoids produce calcite mesocrystals.
Biao Jin, Chunhui Wu, Thi Kim Hoang Trinh, Wenhao Zhou, Chun‐Long Chen, James J De Yoreo   +5 more
wiley   +1 more source