Matches in SemOpenAlex for { <https://semopenalex.org/work/W3097754836> ?p ?o ?g. }
- W3097754836 abstract "A regular convolution layer applying a filter in the same way over known and unknown areas causes visual artifacts in the inpainted image. Several studies address this issue with feature re-normalization on the output of the convolution. However, these models use a significant amount of learnable parameters for feature re-normalization, or assume a binary representation of the certainty of an output. We propose (layer-wise) feature imputation of the missing input values to a convolution. In contrast to learned feature re-normalization, our method is efficient and introduces a minimal number of parameters. Furthermore, we propose a revised gradient penalty for image inpainting, and a novel GAN architecture trained exclusively on adversarial loss. Our quantitative evaluation on the FDF dataset reflects that our revised gradient penalty and alternative convolution improves generated image quality significantly. We present comparisons on CelebA-HQ and Places2 to current state-of-the-art to validate our model." @default.
- W3097754836 created "2020-11-09" @default.
- W3097754836 creator A5002687932 @default.
- W3097754836 creator A5038367345 @default.
- W3097754836 creator A5048393027 @default.
- W3097754836 date "2020-11-02" @default.
- W3097754836 modified "2023-09-23" @default.
- W3097754836 title "Image Inpainting with Learnable Feature Imputation" @default.
- W3097754836 cites W1901129140 @default.
- W3097754836 cites W1967577110 @default.
- W3097754836 cites W1977398717 @default.
- W3097754836 cites W1999360130 @default.
- W3097754836 cites W2096461046 @default.
- W3097754836 cites W2099471712 @default.
- W3097754836 cites W2100415658 @default.
- W3097754836 cites W2105038642 @default.
- W3097754836 cites W2115273023 @default.
- W3097754836 cites W2146337213 @default.
- W3097754836 cites W2194775991 @default.
- W3097754836 cites W2295936755 @default.
- W3097754836 cites W2548275288 @default.
- W3097754836 cites W2557414982 @default.
- W3097754836 cites W2611104282 @default.
- W3097754836 cites W2614221804 @default.
- W3097754836 cites W2616704603 @default.
- W3097754836 cites W2732026016 @default.
- W3097754836 cites W2738588019 @default.
- W3097754836 cites W2763421725 @default.
- W3097754836 cites W2792263949 @default.
- W3097754836 cites W2796286534 @default.
- W3097754836 cites W2798365772 @default.
- W3097754836 cites W2896434438 @default.
- W3097754836 cites W2907097116 @default.
- W3097754836 cites W2921963407 @default.
- W3097754836 cites W2962760235 @default.
- W3097754836 cites W2962785568 @default.
- W3097754836 cites W2962879692 @default.
- W3097754836 cites W2963231084 @default.
- W3097754836 cites W2963255313 @default.
- W3097754836 cites W2963270367 @default.
- W3097754836 cites W2963330667 @default.
- W3097754836 cites W2963373786 @default.
- W3097754836 cites W2963420272 @default.
- W3097754836 cites W2963521568 @default.
- W3097754836 cites W2963653779 @default.
- W3097754836 cites W2963783084 @default.
- W3097754836 cites W2963890275 @default.
- W3097754836 cites W2963981733 @default.
- W3097754836 cites W2964121744 @default.
- W3097754836 cites W2981189092 @default.
- W3097754836 cites W2981682056 @default.
- W3097754836 cites W2982340830 @default.
- W3097754836 cites W2982763192 @default.
- W3097754836 cites W2985764327 @default.
- W3097754836 cites W2989505547 @default.
- W3097754836 cites W3035574324 @default.
- W3097754836 cites W3043547428 @default.
- W3097754836 cites W345598540 @default.
- W3097754836 hasPublicationYear "2020" @default.
- W3097754836 type Work @default.
- W3097754836 sameAs 3097754836 @default.
- W3097754836 citedByCount "0" @default.
- W3097754836 crossrefType "posted-content" @default.
- W3097754836 hasAuthorship W3097754836A5002687932 @default.
- W3097754836 hasAuthorship W3097754836A5038367345 @default.
- W3097754836 hasAuthorship W3097754836A5048393027 @default.
- W3097754836 hasConcept C115961682 @default.
- W3097754836 hasConcept C11727466 @default.
- W3097754836 hasConcept C119857082 @default.
- W3097754836 hasConcept C136886441 @default.
- W3097754836 hasConcept C138885662 @default.
- W3097754836 hasConcept C144024400 @default.
- W3097754836 hasConcept C153180895 @default.
- W3097754836 hasConcept C154945302 @default.
- W3097754836 hasConcept C19165224 @default.
- W3097754836 hasConcept C2776401178 @default.
- W3097754836 hasConcept C33923547 @default.
- W3097754836 hasConcept C41008148 @default.
- W3097754836 hasConcept C41895202 @default.
- W3097754836 hasConcept C45347329 @default.
- W3097754836 hasConcept C48372109 @default.
- W3097754836 hasConcept C50644808 @default.
- W3097754836 hasConcept C58041806 @default.
- W3097754836 hasConcept C9357733 @default.
- W3097754836 hasConcept C94375191 @default.
- W3097754836 hasConceptScore W3097754836C115961682 @default.
- W3097754836 hasConceptScore W3097754836C11727466 @default.
- W3097754836 hasConceptScore W3097754836C119857082 @default.
- W3097754836 hasConceptScore W3097754836C136886441 @default.
- W3097754836 hasConceptScore W3097754836C138885662 @default.
- W3097754836 hasConceptScore W3097754836C144024400 @default.
- W3097754836 hasConceptScore W3097754836C153180895 @default.
- W3097754836 hasConceptScore W3097754836C154945302 @default.
- W3097754836 hasConceptScore W3097754836C19165224 @default.
- W3097754836 hasConceptScore W3097754836C2776401178 @default.
- W3097754836 hasConceptScore W3097754836C33923547 @default.
- W3097754836 hasConceptScore W3097754836C41008148 @default.
- W3097754836 hasConceptScore W3097754836C41895202 @default.
- W3097754836 hasConceptScore W3097754836C45347329 @default.
- W3097754836 hasConceptScore W3097754836C48372109 @default.