Video processing inside embedded devices using SSD-Mobilenet to count mobility actors

Heredia, Andrés; Barros-Gavilanes, Gabriel

doi:10.1109/colcaci.2019.8781798

Cited by 26 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MobileNet achieved these results with an efficient computational performance in mobile devices due to their light-weight architecture. Su et al [20] and Heredia et al [21] reported that it was possible to process between 6 and 10 frames per second with MobileNet architectures in constrained devices.…”

Section: Resultsmentioning

confidence: 99%

Lung ultrasound for point-of-care COVID-19 pneumonia stratification: computer-aided diagnostics in a smartphone. First experiences classifying semiology from public datasets

Almeida

Bilbao-Jayo

Ruby

et al. 2020

2020 IEEE International Ultrasonics Symposium (IUS)

View full text Add to dashboard Cite

Lung ultrasound (LUS) has demonstrated potential in managing pneumonia patients, and is actively used at the point-of-care in COVID-19 patient stratification. However, image interpretation is presently both time-consuming and operatordependent. We explore computer-aided diagnostics of pneumonia semiology based on light-weight neural networks (MobileNets). For proof-of-concept, multi-task learning is performed from online available COVID-19 datasets, for which semiology (overall abnormality, B-lines, consolidations and pleural thickening) is annotated by two radiologists. Initial results suggest that individual indications can be classified with good performance in a smartphone. Neural networks may also help to reduce interreader variability and objectivize LUS interpretation, especially for early-stage pathological indications.

show abstract

Section: Resultsmentioning

confidence: 99%

Lung ultrasound for point-of-care COVID-19 pneumonia stratification: computer-aided diagnostics in a smartphone. First experiences classifying semiology from public datasets

Almeida

Bilbao-Jayo

Ruby

et al. 2020

2020 IEEE International Ultrasonics Symposium (IUS)

View full text Add to dashboard Cite

show abstract

“…The information is passed to the user via audio output. Object detection may be possible by having a pre-trained deep learning model such as YoloV3 [47] or MobileNet+SSDv2 [22] in the smartphone itself. The depth camera can be utilised to determine the distance of the object from the user from the video frames itself.…”

Section: Usecase 1: Object Detection and Obstacle Avoidancementioning

confidence: 99%

Smartphone Navigation Support for Blind and Visually Impaired People - A Comprehensive Analysis of Potentials and Opportunities

Kuriakose

Shrestha

Sandnes

2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Smartphones are indispensable tools for many people for various assistive tasks. They are equipped with different sensors that provide data on motion, location and the environment. This paper explores the various sensors and output modalities existing in a smartphone to make it useful as a navigation support device for blind and visually impaired users. In addition, different usecase scenarios were also scrutinized where the potential of a smartphone as a navigation device can be explored further. The technology holds potential for the implementation of successful navigation support system by utilizing the various sensors and features existing in a smartphone. This in-depth analysis of the various possibilities with a smartphone in the navigation support system design might become useful in the further research in the domain.

show abstract

“…We use a pretrained full YOLOv3 model as the teacher, and train smaller, specialized detectors for each camera. The specialized models are built on SSD-MobileNets and can be deployed on embedded devices [40], [13]. Specialized detectors are covered in recent approaches; we focus on the identification layers.…”

Section: B Teamed Classifiers For Vehicle Re-idmentioning

confidence: 99%

Robust, Extensible, and Fast: Teamed Classifiers for Vehicle Tracking in Multi-Camera Networks

Suprem

Lima

Padilha

et al. 2019

2019 IEEE First International Conference on Cognitive Machine Intelligence (CogMI)

View full text Add to dashboard Cite

As camera networks have become more ubiquitous over the past decade, the research interest in video management has shifted to analytics on multi-camera networks. This includes performing tasks such as object detection, attribute identification, and vehicle/person tracking across different cameras without overlap. Current frameworks for management are designed for multi-camera networks in a closed dataset environment where there is limited variability in cameras and characteristics of the surveillance environment are well known. Furthermore, current frameworks are designed for offline analytics with guidance from human operators for forensic applications. This paper presents a teamed classifier framework for video analytics in heterogeneous many-camera networks with adversarial conditions such as multi-scale, multi-resolution cameras capturing the environment with varying occlusion, blur, and orientations. We describe an implementation for vehicle tracking and vehicle re-identification (re-id), where we implement a zero-shot learning (ZSL) system that performs automated tracking of all vehicles all the time. Our evaluations on VeRi-776 and Cars196 show the teamed classifier framework is robust to adversarial conditions, extensible to changing video characteristics such as new vehicle types/brands and new cameras, and offers real-time performance compared to current offline video analytics approaches.

show abstract

Video processing inside embedded devices using SSD-Mobilenet to count mobility actors

Cited by 26 publications

References 8 publications

Lung ultrasound for point-of-care COVID-19 pneumonia stratification: computer-aided diagnostics in a smartphone. First experiences classifying semiology from public datasets

Lung ultrasound for point-of-care COVID-19 pneumonia stratification: computer-aided diagnostics in a smartphone. First experiences classifying semiology from public datasets

Smartphone Navigation Support for Blind and Visually Impaired People - A Comprehensive Analysis of Potentials and Opportunities

Robust, Extensible, and Fast: Teamed Classifiers for Vehicle Tracking in Multi-Camera Networks

Contact Info

Product

Resources

About