prikaz prve stranice dokumenta Efficient semantic image segmentation using pyramidal fusion
  Download
PDF 41.67 MB
doctoral thesis
Efficient semantic image segmentation using pyramidal fusion
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. urn:nbn:hr:168:672658
Oršić, Marin
University of Zagreb
Faculty of Electrical Engineering and Computing
Department of Electronics, Microelectronics, Computer and Intelligent Systems

Institutional repository: FER Repository

Cite this document

Oršić, M. (2021). Efficient semantic image segmentation using pyramidal fusion (Doctoral thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:672658

Oršić, Marin. "Efficient semantic image segmentation using pyramidal fusion." Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:672658

Oršić, Marin. "Efficient semantic image segmentation using pyramidal fusion." Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:672658

Oršić, M. (2021). 'Efficient semantic image segmentation using pyramidal fusion', Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 02 June 2024, https://urn.nsk.hr/urn:nbn:hr:168:672658

Oršić M. Efficient semantic image segmentation using pyramidal fusion [Doctoral thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2021 [cited 2024 June 02] Available at: https://urn.nsk.hr/urn:nbn:hr:168:672658

M. Oršić, "Efficient semantic image segmentation using pyramidal fusion", Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:168:672658

Metadata
TitleEfficient semantic image segmentation using pyramidal fusion
Title (croatian)Učinkovita semantička segmentacija slike piramidnom fuzijom
AuthorMarin Oršić
MentorSiniša Šegvić (mentor)
Committee memberSiniša Šegvić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
(Department of Electronics, Microelectronics, Computer and Intelligent Systems)
Zagreb
Defense date and country2021-11-12, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Universal decimal classification
(UDC)		004 - Computer science and technology. Computing. Data processing
Abstract
Emergence of large datasets and resilience of convolutional models have enabled successful training of very large semantic segmentation models. However, high capacity implies high computational complexity and therefore hinders real-time operation. We therefore study compact architectures which aim at high accuracy in spite of modest capacity. We propose a novel semantic segmentation approach based on shared pyramidal representation and fusion of heterogeneous features along the upsampling path. The proposed pyramidal fusion approach is especially effective for dense inference in images with large scale variance due to strong regularization effects induced by feature sharing across the resolution pyramid. Interpretation of the decision process suggests that our approach succeeds by acting as a large ensemble of relatively simple models, as well as due to large receptive range and strong gradient flow towards early layers. Validation and ablation experiments support our design choices and suggest that the proposed approach succeeds by acting as an ensemble of relatively simpler models. Our best model achieves 76.4\% mIoU on Cityscapes test and runs in real time on low-power embedded devices. In this thesis, we describe the main components of a real-time semantic segmentation system based on deep convolutional models. We are considered with convolutional encoders used for recognition, as well as decoders which are crucial for obtaining accurate results. We do extensive evaluation of the developed method over a range of public and domestic datasets. Finally, we present results in the 2020 instance of Robust Vision Challenge.
Abstract (croatian)
Pojava velikih podatkovnih skupova i otpornost konvolucijskih modela omogućili su uspješno treniranje velikih modela za semantičku segmentaciju. Doduše, velik kapacitet modela uzorkuje visoku računsku složenost što onemogućuje primjene u stvarnom vremenu. Razmatramo kompaktne arhitekture koje ganjaju visoku prediktivnu moć usprkos skromnom kapacitetu. Predlažemo novi pristup za semantičku segmentaciju slika temeljenu na spajanju dijeljenih heterogenih reprezentacija izlučenih iz rezolucijske piramide. Predloženi pristup, zvan piramidna fuzija, je naročito učinkovit u gustoj predikciji na slikama velike varijance u skali zbog snažnog regularizacijskog učinka potaknutog dijeljenjem značajki preko rezolucijske piramide. Interpretacija toka donošenja odluke naslućuje kako se naš pristup ponaša kao ansambl jednostavnijih modela, te kako široko vidno polje i jak tok gradijenata pridonose visokoj točnosti. Validacijski i ablacijski eksperimenti podržavaju naše odluke i sugeriraju kako se predloženi pristup uspješno ponaša kao ansambl relativno jednostavnijih modela. Naš najbolji model postiže 76.4\% mIoU na ispitnom podskupu podatkovnog skupa Cityscapes te podržava izvođenje u stvarnom vremenu na ugradbenim računalima niske potrošnje. Ova disertacija opisuje glavne dijelove sustava za semantičku segmentaciju slika temeljenog na dubokim konvolucijskim modelima. Razmatramo konvolucijske kodere za raspoznavanje te dekodere koji su neizostavni u ostvarivanju točnih rezultata. Pokazujemo obilan skup evaluacija razvijene metode u nizu javno dostupnih i domaćih podatkovnih skupova. Konačno, prikazujemo rezultate našeg sudjelovanja u natjecanju Robust Vision Challenge 2020.
Keywords
Keywords (croatian)
Languageenglish
URN:NBNurn:nbn:hr:168:672658
Promotion2022
Study programmeTitle: Computer Science
Study programme type: university
Study level: postgraduate
Academic / professional title: Doktor znanosti (Doktor znanosti)
Type of resourceText
Extentviii, 84 str. : graf. prikazi
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2022-07-05 08:53:44