FedViT: Federated continual learning of vision transformer at edge

Deep Neural Networks (DNNs) have been ubiquitously adopted in internet of things and are becoming an integral part of our daily life. When tackling the evolving learning tasks in real world, such as classifying different types of objects, DNNs face the challenge to continually retrain themselves according to the tasks on different edge...

Efficient Temporal Action Localization via Vision-Language Modelling: An Empirical Study on the STALE Model's Efficiency and Generalizability in Resource-constrained Environments

Temporal Action Localization (TAL) aims to localize the start and end times of actions in untrimmed videos and classify the corresponding action types. TAL plays an important role in understanding video. Existing TAL approaches heavily rely on deep learning and require large-scale data and expensive training processes. Recent advances in...

EdgeVisionBench: A Benchmark of Evolving Input Domains for Vision Applications at Edge

Vision applications powered by deep neural networks (DNNs) are widely deployed on edge devices and solve the learning tasks of incoming data streams whose class label and input feature continuously evolve, known as domain shift. Despite its prominent presence in real-world edge scenarios, existing benchmarks used by domain adaptation methods...

On-Line Black-Box Aerodynamic Performance Optimization for a Morphing Wing With Distributed Sensing and Control

Inspired by nature, smart morphing technologies enable the aircraft of tomorrow to sense their environment and adapt the shape of their wings in flight to minimize fuel consumption and emissions. A primary challenge on the road to this feature is how to use the knowledge gathered from sensory data to establish an optimal shape adaptively and...

Vision-Based Nonlinear Incremental Control for A Morphing Wing with Mechanical Imperfections

Morphing structures have acquired much attention in the aerospace community because they enable an aircraft to actively adapt its shape during flight, leading to fewer emissions and fuel consumption. Researchers have designed, manufactured, and tested a morphing wing named SmartX-Alpha, which can actively alleviate loads while achieving the...

Binary Neural Networks for Object Detection

In the past few years, convolutional neural networks (CNNs) have been widely utilized and shown state-of-the-art performances on computer vision tasks. However, CNN based approaches usually require a large amount of storage, run-time memory, as well as computation power in both training and inference time, which are usually used on GPU based...

Depth-aware Instance Segmentation with a Discriminative Loss Function

This work explores the possibility of incorporating depth information into a deep neural network to improve accuracy of RGB instance segmentation. The baseline of this work is semantic instance segmentation with discriminative loss function.The baseline work proposes a novel discriminative loss function with which the semantic net-work can learn...

High-Performance Back-Illuminated Three-Dimensional Stacked Single-Photon Avalanche Diode Implemented in 45-nm CMOS Technology

We present a high-performance back-illuminated three-dimensional stacked single-photon avalanche diode (SPAD), which is implemented in 45-nm CMOS technology for the first time. The SPAD is based on a P⁺/Deep N-well junction with a circular shape, for which N-well is intentionally excluded to achieve a wide depletion region, thus...

Active Vision for Humanoid Robots

Human perception is an active process. By altering its viewpoint rather than passively observing surroundings and by operating on sequences of images rather than on a single frame, the human visual system has the ability to explore the most relevant information based on knowledge, therefore when growing up a human is able to develop cognitive...