Pregled bibliografske jedinice broj: 1067520
Perceptual Autoencoder for Compressive Sensing Image Reconstruction
Perceptual Autoencoder for Compressive Sensing Image Reconstruction // Informatica, 1 (2020), 1; 1-18 doi:10.15388/20-infor421 (međunarodna recenzija, članak, znanstveni)
CROSBI ID: 1067520 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Perceptual Autoencoder for Compressive Sensing
Image Reconstruction
Autori
Ralašić, Ivan ; Seršić, Damir ; Šegvić, Siniša
Izvornik
Informatica (0868-4952) 1
(2020), 1;
1-18
Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni
Ključne riječi
compressive sensing ; convolutional autoencoder ; deep learning ; image reconstruction ; perceptual loss ; principal component analysis
Sažetak
This paper presents a non-iterative deep learning approach to compressive sensing (CS) image reconstruction using a convolutional autoencoder and a residual learning network. An efficient measurement design is proposed in order to enable training of the compressive sensing models on normalized and mean-centred measurements, along with a practical network initialization method based on principal component analysis (PCA). Finally, perceptual residual learning is proposed in order to obtain semantically informative image reconstructions along with high pixel-wise reconstruction accuracy at low measurement rates.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
Projekti:
HRZZ-IP-2019-04-6703 - Renesansa teorije uzorkovanja (SamplingRenaissance) (Seršić, Damir, HRZZ ) ( CroRIS)
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb
Citiraj ovu publikaciju:
Časopis indeksira:
- Current Contents Connect (CCC)
- Web of Science Core Collection (WoSCC)
- Science Citation Index Expanded (SCI-EXP)
- SCI-EXP, SSCI i/ili A&HCI
- Scopus
