Detecting Objects for Autonomous System Verificatio
Result description
In this thesis we created a framework for easy evaluation and training of Faster R-CNN type of networks. We fine-tuned VGG16 and ZFNet networks on our internal Victims dataset as well as standard KITTI dataset. We later showed that VGG16 architecture is far more suitable for fine-tuning on data from slightly different training and target domains. This framework can later serve as a baseline for further improvements in the field.
Keywords
The result's identifiers
Result code in IS VaVaI
Result on the web
https://dspace.cvut.cz/bitstream/handle/10467/64667/F3-BP-2016-Jasek-Otakar-jasek.pdf
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Detecting Objects for Autonomous System Verificatio
Original language description
In this thesis we created a framework for easy evaluation and training of Faster R-CNN type of networks. We fine-tuned VGG16 and ZFNet networks on our internal Victims dataset as well as standard KITTI dataset. We later showed that VGG16 architecture is far more suitable for fine-tuning on data from slightly different training and target domains. This framework can later serve as a baseline for further improvements in the field.
Czech name
—
Czech description
—
Classification
Type
Vsouhrn - Summary research report
CEP classification
JD - Use of computers, robotics and its application
OECD FORD branch
—
Result continuities
Project
—
Continuities
R - Projekt Ramcoveho programu EK
Others
Publication year
2016
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Number of pages
60
Place of publication
Praha
Publisher/client name
ČVUT FEL, Katedra kybernetiky - Centrum strojového vnímání
Version
—
Basic information
Result type
Vsouhrn - Summary research report
CEP
JD - Use of computers, robotics and its application
Year of implementation
2016