# A Light-powered, "Always-On", Smart Camera with Compressed Domain Gesture Detection

@article{Anvesha2016AL,
title={A Light-powered, "Always-On", Smart Camera with Compressed Domain Gesture Detection},
author={Amaravati Anvesha and Shaojie Xu and Ningyuan Cao and Justin K. Romberg and Arijit Raychowdhury},
journal={Proceedings of the 2016 International Symposium on Low Power Electronics and Design},
year={2016}
}
• A. Anvesha, +2 authors A. Raychowdhury
• Published 26 May 2016
• Computer Science
• Proceedings of the 2016 International Symposium on Low Power Electronics and Design
In this paper we propose an energy-efficient camera-based gesture recognition system powered by light energy for "always on" applications. Low energy consumption is achieved by directly extracting gesture features from the compressed measurements, which are the block averages and the linear combinations of the image sensor's pixel values. The gestures are recognized using a nearest-neighbour (NN) classifier followed by Dynamic Time Warping (DTW). The system has been implemented on an Analog…
9 Citations
A Light-Powered Smart Camera With Compressed Domain Gesture Detection
• Computer Science
IEEE Transactions on Circuits and Systems for Video Technology
• 2018
Owing to its fully self-powered operation, the proposed system can find wide applications in “always-on” vision systems, such as in surveillance, robotics, and consumer electronics with touch-less operation.
Appearance-based gesture recognition in the compressed domain
• Computer Science, Mathematics
2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
• 2017
This work proposes a novel appearance-based gesture recognition algorithm using compressed domain signal processing techniques that improves both the computational efficiency and the memory requirement of the previous DTW-based K-NN gesture classifiers.
A 130 nm 165 nJ/frame Compressed-Domain Smashed-Filter-Based Mixed-Signal Classifier for “In-Sensor” Analytics in Smart Cameras
• Computer Science
IEEE Transactions on Circuits and Systems II: Express Briefs
• 2018
A compressed-domain smashed-filter-based object recognition system and measurements from a 130 nm mixed-signal test chip are proposed to demonstrate reconfigurable and ultra-low power operation.
Self-Optimizing IoT Wireless Video Sensor Node With In-Situ Data Analytics and Context-Driven Energy-Aware Real-Time Adaptation
• Computer Science
IEEE Transactions on Circuits and Systems I: Regular Papers
• 2017
This paper presents a camera-based wireless sensor node with a self-optimizing end-to-end computation and communication design, targeted for surveillance applications, and demonstrates support for multiple feature-extraction and classification algorithms, tunable processing depth and power amplifier gain.
Charge for a whole day: Extending Battery Life for BCI Wearables using a Lightweight Wake-Up Command
• Computer Science
CHI
• 2020
Trance is presented, a user-friendly and robust wake-up command for BCI headsets that is computationally lightweight and can extend battery life by approximately 2.7x or to approximately 10 hours for a typical wearable battery, while remaining user- friendly.
In-sensor analytics and energy-aware self-optimization in a wireless sensor node
• Engineering
2017 IEEE MTT-S International Microwave Symposium (IMS)
• 2017
With the proliferation of distributed sensors and Internet of Thing end-nodes, aggregate data transfer to the back-end servers in the cloud is expected to become prohibitively large which not only
Challenges of the digital transformation: The role of sensors, sensor networks, IoT-devices, and 5G : Invited Paper
• Computer Science
2019 First International Conference on Societal Automation (SA)
• 2019
The importance of sensors and sensing concepts for digitalization of large automation systems and recent component developments, standards and systematic ideas, which may be relevant for large sensor systems in many areas are described.
Sensors as the Basis for Digitalization: New Approaches in Instrumentation, IoT-concepts, and 5G
• Computer Science
Internet Things
• 2021
The role of sensors and sensing concepts is reviewed and reflected in the context of digitalization, particularly of large automation systems, and specific 5G-related opportunities for new sensor applications are discussed.
Context-Aware Intelligence in Resource-Constrained IoT Nodes: Opportunities and Challenges
• Computer Science
IEEE Design & Test
• 2019
An academic perspective of the problem is provided, starting with a survey of recent advances in intelligent sensing, computation, communication, and energy management for resource-constrained IoT sensor nodes and leading to a future outlook and needs.

## References

SHOWING 1-10 OF 17 REFERENCES
An Ultra-Low Power, “Always-On” Camera Front-End for Posture Detection in Body Worn Cameras Using Restricted Boltzman Machines
• Engineering, Computer Science
IEEE Transactions on Multi-Scale Computing Systems
• 2015
An RBM based “always-on’’ camera front-end capable of detecting human posture and Aggressive behavior of the human being in the field of view will be used as a wake-up signal for further data collection and classification.
Real-time visual static hand gesture recognition system and its FPGA-based hardware implementation
• 2014 12th International Conference on Signal Processing (ICSP)
• 2014
Gesture recognition has been paid more and more attention as a new generation of human-computer interaction and visual input mode. In motion sensing games and other applications, gesture recognition
Real-time visual static hand gesture recognition system and its FPGA-based hardware implementation
• Computer Science
• 2014
This paper optimizes the existing fingertip detection algorithms with the proposed essential constraints, which improves the recognition rate to 90.3% and the operating speed is improved significantly.
An FPGA-Based Smart Camera for Gesture Recognition in HCI Applications
• Computer Science
ACCV
• 2007
GestureCam is presented, an FPGA-based smart camera built from scratch that can recognize simple hand gestures and has shown promising real-time performance and is being tested in several desktop HCI (Human Computer Interface) applications.
A Novel FPGA-based Hand Gesture Recognition System
• Computer Science
• 2012
A hardware implemented gesture recognition system based on improved optical flow was proposed to accelerate the execution and achieve real-time operation and results showed that the proposed framework is feasible and can reach the speed of 30 frames per second.
Vision-based remote control system by motion detection and open finger counting
• Computer Science
IEEE Transactions on Consumer Electronics
• 2009
Experimental results show that the shape of open fingers exhibits strong features for determining correct gesture states and the use of stable features on consecutive frames yields robust and accurate performance regardless of operating conditions.
A Batteryless 19 $\mu$W MICS/ISM-Band Energy Harvesting Body Sensor Node SoC for ExG Applications
• Yanqing Zhang, +10 authors B. Otis
• Engineering, Computer Science
IEEE Journal of Solid-State Circuits
• 2013
This SoC is designed so the integration and interaction of circuit blocks accomplish an integrated, flexible, and reconfigurable wireless BSN SoC capable of autonomous power management and operation from harvested power, thus prolonging the node lifetime indefinitely.
Vision based hand gesture recognition for human computer interaction: a survey
• Computer Science
Artificial Intelligence Review
• 2012
An analysis of comparative surveys done in the field of gesture based HCI and an analysis of existing literature related to gesture recognition systems for human computer interaction by categorizing it under different key parameters are provided.
A Highly Efficient Ultralow Photovoltaic Power Harvesting System With MPPT for Internet of Things Smart Nodes
• Engineering, Computer Science
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
• 2015
A hill-climbing maximum power point tracking algorithm is developed in an energy-efficient manner to tune the input impedance of the system and guarantee adaptive maximum power transfer under wide illumination conditions, and a capacitor value modulation is implemented to achieve a higher efficiency than the traditional pulse-frequency modulation scheme.
The smashed filter for compressive classification and target recognition
A framework for compressive classification that operates directly on the compressive measurements without first reconstructing the image is proposed, and the effectiveness of the smashed filter for target classification using very few measurements is demonstrated.