3 papers
cs.CV2023
FPGA-QHAR: Throughput-Optimized for Quantized Human Action Recognition on The Edge
Azzam Alhussain, Mingjie Lin
Accelerating Human Action Recognition (HAR) efficiently for real-time surveillance and robotic systems on edge chips remains a challenging research field, given its high computatio…
cs.AR2022
Hardware-Efficient Template-Based Deep CNNs Accelerator Design
Azzam Alhussain, Mingjie Lin
Acceleration of Convolutional Neural Network (CNN) on edge devices has recently achieved a remarkable performance in image classification and object detection applications. This pa…
cs.LG2022
Hardware-Efficient Deconvolution-Based GAN for Edge Computing
Azzam Alhussain, Mingjie Lin
Generative Adversarial Networks (GAN) are cutting-edge algorithms for generating new data samples based on the learned data distribution. However, its performance comes at a signif…