Imageinpaintingwithlearnablefeatureimputation

2020年11月2日 — We propose (layer-wise) feature imputation of the missing input values to a convolution. In contrast to learned feature re-normalization, our ...

由 H Hukkelås 著作 · 2020 · 被引用 14 次 — We propose (layer-wise) feature imputation of the missing input values to a convolution. In contrast to learned feature re-normalization, our ...

The following items are included in the supplementary material: 1. Appendix 1 includes experiments analysing the impact of L∞ vs L2 norm for gra-.

This chapter proposes (layer-wise) feature imputation of the missing input values to a convolution, and proposes a revised gradient penalty for image ...