Detecting Wearable Objects via Transfer Learning

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F19%3A00337709" target="_blank" >RIV/68407700:21730/19:00337709 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1109/ICCP48234.2019.8959621" target="_blank" >https://doi.org/10.1109/ICCP48234.2019.8959621</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICCP48234.2019.8959621" target="_blank" >10.1109/ICCP48234.2019.8959621</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Detecting Wearable Objects via Transfer Learning

Popis výsledku v původním jazyce

Transfer learning is a well known technique to circumvent the problem of small datasets in deep machine learning. It has been successfully used in the field of camera surveillance image processing which suffers from poor data quality and quantity. We focused on the task of wearable object detection, namely distinguishing if a person is or is not wearing a backpack. We created new annotations for the DukeMTMC-attribute dataset to overcome the discrepancies among the attributes. We explored transfer learning with a frozen feature extractor as well as the model fine-tuning, which turned out to perform much better. In both setups we found that the Densenet161 is the best from tested architectures. Our best model achieved about 92% balanced accuracy on the testing set.

Název v anglickém jazyce

Detecting Wearable Objects via Transfer Learning

Popis výsledku anglicky

Transfer learning is a well known technique to circumvent the problem of small datasets in deep machine learning. It has been successfully used in the field of camera surveillance image processing which suffers from poor data quality and quantity. We focused on the task of wearable object detection, namely distinguishing if a person is or is not wearing a backpack. We created new annotations for the DukeMTMC-attribute dataset to overcome the discrepancies among the attributes. We explored transfer learning with a frozen feature extractor as well as the model fine-tuning, which turned out to perform much better. In both setups we found that the Densenet161 is the best from tested architectures. Our best model achieved about 92% balanced accuracy on the testing set.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20202 - Communication engineering and systems

Projekt

<a href="/cs/project/VI20172019082" target="_blank" >VI20172019082: Smart Camera - Dohledové centrum nové generace</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2019

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

2019 IEEE 15th International Conference on Intelligent Computer Communication and Processing (ICCP)

ISBN

978-1-7281-4914-1

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

373-380

Název nakladatele

IEEE

Místo vydání

Piscataway, NJ

Místo konání akce

Cluj-Napoca

Datum konání akce

5. 9. 2019

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—