[HTML][HTML] 7 μJ/inference end-to-end gesture recognition from dynamic vision sensor data using ternarized hybrid convolutional neural networks

G Rutishauser, M Scherer, T Fischer, L Benini - Future Generation …, 2023 - Elsevier
Dynamic vision sensor (DVS) cameras enable energy-activity proportional visual sensing by
only propagating events produced by changes in the observed scene. Furthermore, by …

TCN-CUTIE: A 1,036-TOp/s/W, 2.72-µJ/Inference, 12.2-mW All-Digital Ternary Accelerator in 22-nm FDX Technology

M Scherer, A Di Mauro, T Fischer, G Rutishauser… - IEEE Micro, 2022 - ieeexplore.ieee.org
Tiny machine learning (TinyML) applications impose µJ/inference constraints, with a
maximum power consumption of tens of megawatt. It is extremely challenging to meet these …

A 593nJ/Inference DVS Hand Gesture Recognition Processor Embedded With Reconfigurable Multiple Constant Multiplication Technique

Z Fu, W Ye - IEEE Transactions on Circuits and Systems I …, 2024 - ieeexplore.ieee.org
Hand gesture recognition (HGR) is a popular technique for edge-based human-computer
interaction. Dynamic vision sensors (DVS) are often used in HGR systems due to their low …

Agile and Efficient Inference of Quantized Neural Networks

G Rutishauser - 2024 - research-collection.ethz.ch
Zeitgleich mit der rasanten Ausbreitung des Internet of Things (IoT) hat die Entwicklung von
Deep-Learning-Algorithm eine Revolution im Feld des maschinellen Lernens ausgelöst. Die …

[PDF][PDF] Hardware-Software Co-Design for Energy-Efficient Neural Network Inference at the Extreme Edge

M Scherer - 2024 - research-collection.ethz.ch
Since the breakthrough success of AlexNet in the ILSVRC image recognition challenge in
2012, Deep Neural Networks (DNNs), and in particular Convolutional Neural Networks …