Free-Form Image Inpainting with Gated Convolution — mentioned by zawerf

Mon, 29 Oct 2018 17:35:21 GMT

I believe "Image Inpainting for Irregular Holes Using Partial Convolutions" has been superseded by "Free-Form Image Inpainting with Gated Convolution", especially for interactive use cases? https://arxiv.org/pdf/1806.03589.pdf http://jiahuiyu.com/deepfill2/ I think the most interesting part of this project would be generating the dataset. That is a lot of manual redrawing of vaginas/penises. Or the reverse, where you find already decensored stuff and add realistic censorship yourself. Either way, a lot of time will be spent tagging genitals to build a large enough dataset to make NN…

Paper and discussion history. A mention is not an endorsement.

Free-Form Image Inpainting with Gated Convolution — mentioned by mattgroh

Mon, 03 Sep 2018 15:53:31 GMT

If you're interested in seeing the top of the line inpainting results, you should check out DeepFill and DeepFill2. http://jiahuiyu.com/deepfill2/ Deep Angel is based on an architecture combining Mask R-CNN (for object detection and instance segementation) and DeepFill for image inpainting. And here's the paper behind DeepFill https://arxiv.org/pdf/1806.03589.pdf If you look at the papers, you'll see which one has the best inpainting results.

Paper and discussion history. A mention is not an endorsement.

Free-Form Image Inpainting with Gated Convolution · HackerPapers discussions

Free-Form Image Inpainting with Gated Convolution — mentioned by zawerf

Free-Form Image Inpainting with Gated Convolution — mentioned by mattgroh