A Generative Adversarial Framework for Optimizing Image Matting and Harmonization Simultaneously

Xuqian Ren, Yifan Liu, Chunlei Song

Tutkimustuotos: KonferenssiartikkeliTieteellinenvertaisarvioitu

6 Sitaatiot (Scopus)
8 Lataukset (Pure)

Abstrakti

Image matting and image harmonization are two important tasks in image composition. Image matting, aiming to achieve foreground boundary details, and image harmonization, aiming to make the background compatible with the foreground, are both promising yet challenging tasks. Previous works consider optimizing these two tasks separately, which may lead to a sub-optimal solution. We propose to optimize matting and harmonization simultaneously to get better performance on both the two tasks and achieve more natural results. We propose a new Generative Adversarial (GAN) framework which optimizing the matting network and the harmonization network based on a self-attention discriminator. The discriminator is required to distinguish the natural images from different types of fake synthesis images. Extensive experiments on our constructed dataset demonstrate the effectiveness of our proposed method. Our dataset and dataset generating pipeline can be found in \url{https://git.io/HaMaGAN}
AlkuperäiskieliEnglanti
Otsikko2021 IEEE International Conference on Image Processing (ICIP)
KustantajaIEEE
Sivut1354-1358
ISBN (elektroninen)978-1-6654-4115-5
DOI - pysyväislinkit
TilaJulkaistu - 13 elok. 2021
Julkaistu ulkoisestiKyllä
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaIEEE International Conference on Image Processing - , Yhdysvallat
Kesto: 19 syysk. 202122 syysk. 2021

Julkaisusarja

NimiICIP

Conference

ConferenceIEEE International Conference on Image Processing
Maa/AlueYhdysvallat
Ajanjakso19/09/2122/09/21

Sormenjälki

Sukella tutkimusaiheisiin 'A Generative Adversarial Framework for Optimizing Image Matting and Harmonization Simultaneously'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä