Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Multitask Deep Learning models for real-time deployment in embedded systems
KTH, School of Computer Science and Communication (CSC), Robotics, perception and learning, RPL. Universitat Politècnica de Catalunya.
2017 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesisAlternative title
Deep Learning-modeller för multitaskproblem, anpassade för inbyggda system i realtidsapplikationer (Swedish)
Abstract [en]

Multitask Learning (MTL) was conceived as an approach to improve thegeneralization ability of machine learning models. When applied to neu-ral networks, multitask models take advantage of sharing resources forreducing the total inference time, memory footprint and model size. Wepropose MTL as a way to speed up deep learning models for applicationsin which multiple tasks need to be solved simultaneously, which is par-ticularly useful in embedded, real-time systems such as the ones foundin autonomous cars or UAVs.In order to study this approach, we apply MTL to a Computer Vi-sion problem in which both Object Detection and Semantic Segmenta-tion tasks are solved based on the Single Shot Multibox Detector andFully Convolutional Networks with skip connections respectively, usinga ResNet-50 as the base network. We train multitask models for twodifferent datasets, Pascal VOC, which is used to validate the decisionsmade, and a combination of datasets with aerial view images capturedfrom UAVs.Finally, we analyse the challenges that appear during the process of train-ing multitask networks and try to overcome them. However, these hinderthe capacity of our multitask models to reach the performance of the bestsingle-task models trained without the limitations imposed by applyingMTL. Nevertheless, multitask networks benefit from sharing resourcesand are 1.6x faster, lighter and use less memory compared to deployingthe single-task models in parallel, which turns essential when runningthem on a Jetson TX1 SoC as the parallel approach does not fit intomemory. We conclude that MTL has the potential to give superior per-formance as far as the object detection and semantic segmentation tasksare concerned in exchange of a more complex training process that re-quires overcoming challenges not present in the training of single-taskmodels.

Place, publisher, year, edition, pages
2017. , p. 69
Series
EES Examensarbete / Master ThesisEES Examensarbete / Master Thesis
Keywords [en]
computer vision, deep learning, multitask learning, object detection, semantic segmentation, embedded systems, perception, robotics, autonomous driving
National Category
Robotics
Identifiers
URN: urn:nbn:se:kth:diva-216673OAI: oai:DiVA.org:kth-216673DiVA, id: diva2:1151629
External cooperation
Universitat Politècnica de Catalunya; National Institute of Informatics, Japan
Subject / course
Electrical Engineering
Educational program
Master of Science in Engineering - Electrical Engineering
Supervisors
Examiners
Available from: 2017-10-25 Created: 2017-10-23 Last updated: 2017-10-25Bibliographically approved

Open Access in DiVA

fulltext(61348 kB)93 downloads
File information
File name FULLTEXT01.pdfFile size 61348 kBChecksum SHA-512
e7cef8869f5146fb3c385ae849265ed8064574ee5a675f14c3c238383dc3d333690bcb4f6199a6ee2722d6ee2a3082c265f2c8c85b6e7abbca5b769bd9e67383
Type fulltextMimetype application/pdf

By organisation
Robotics, perception and learning, RPL
Robotics

Search outside of DiVA

GoogleGoogle Scholar
Total: 93 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 178 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf