prikaz prve stranice dokumenta Gusto semantičko prognoziranje regresijom značajki
No public access
undergraduate thesis
Gusto semantičko prognoziranje regresijom značajki
2022. urn:nbn:hr:168:382497
Družijanić, Luka
University of Zagreb
Faculty of Electrical Engineering and Computing

Cite this document

Družijanić, L. (2022). Gusto semantičko prognoziranje regresijom značajki (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:382497

Družijanić, Luka. "Gusto semantičko prognoziranje regresijom značajki." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2022. https://urn.nsk.hr/urn:nbn:hr:168:382497

Družijanić, Luka. "Gusto semantičko prognoziranje regresijom značajki." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2022. https://urn.nsk.hr/urn:nbn:hr:168:382497

Družijanić, L. (2022). 'Gusto semantičko prognoziranje regresijom značajki', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 21 January 2025, https://urn.nsk.hr/urn:nbn:hr:168:382497

Družijanić L. Gusto semantičko prognoziranje regresijom značajki [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2022 [cited 2025 January 21] Available at: https://urn.nsk.hr/urn:nbn:hr:168:382497

L. Družijanić, "Gusto semantičko prognoziranje regresijom značajki", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:168:382497

Please login to the repository to save this object to your list.
Metadata
TitleGusto semantičko prognoziranje regresijom značajki
Title (english)Dense semantic forecasting by feature regression
AuthorLuka Družijanić
MentorSiniša Šegvić (mentor)
Committee memberSiniša Šegvić (predsjednik povjerenstva)
Committee memberJosip Šarić (član povjerenstva)
Committee memberMatej Grcić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2022-07-06, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Predviđanje budućnosti bitan je aspekt umjetne inteligencije. Razmatramo semantičko prognoziranje značajki (F2F), bazirano na modelu za semantičku segmentaciju bez lateralnih veza u naduzorkovanju, što omogućuje F2F modelu da radi samo na značajkama najmanje rezolucije. Nadalje, uspoređujemo nekoliko konvolucijskih arhitektura za F2F prognoziranje. Eksperimenti pokazuju da deformabilne konvolucije postižu bolje rezultate od običnih i dilatiranih konvolucija.
Abstract (english)
Predicting the future is an important aspect of artificial intelligence. We explore feature-to-feature (F2F) semantic forecasting, based on a semantic segmentation model without lateral connections in the upsampling path, which allows the F2F model to work with features of lowest resolution. Furthermore, we compare several convolutional architecture for F2F forecasting. Experiments show that deformable convolutions achieve better results than regular and dilated convolutions.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:382497
Study programmeTitle: Computing
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-15 10:02:20