Improving Output Image Quality

---

The Style Images were collect from here: https://commons.wikimedia.org/wiki/Category:Gigapixel_images_from_the_Google_Art_Project

Each image is around 200MB+ in size, so downloading them directly to the Neural-Style directory is advised:

wget https://upload.wikimedia.org/wikipedia/commons/0/0b/Sandro_Botticelli_-_La_nascita_di_Venere_-_Google_Art_Project_-_edited.jpg

wget https://upload.wikimedia.org/wikipedia/commons/e/ed/Canaletto_-_Bucentaur%27s_return_to_the_pier_by_the_Palazzo_Ducale_-_Google_Art_Project.jpg

Due to the Russian characters in the name, the third style image used needs to be downloaded mannually without wget: Александр Андреевич Иванов - Явление Христа народу (Явление Мессии) - Google Art Project.jpg

---

There are multiple different ways to try and improve the quality of the final output image. The most important is getting your parameters right, but you can improve the quality farther using the methods outline below:

1. Repeating your step(s) with the output image as your new content image:

-content_image <output_image>

• Top left first run, top right second run, bottom left third run, and bottom right fourth run.

• The exact same parameters were used for each "run". Each "run" was composed of 7 multiscale/multires.sh steps starting at 640px and ending at 1920px.

• The example image is composed of 224x224 crops from the center of each output image. They were then resized so that the differences were more easily visible. The non-resized crop set can be found here: https://i.imgur.com/wcb8aCl.png

Direct link: https://i.imgur.com/oK16RYA.png

---

---

Waifu2x Comparisons

---

** Waifu2x Projects:**

1. nagadomi's waifu2x

• Supports Linux.

2. waifu2x-caffe

• Supports Windows.

3. WL-Amigo's waifu2x-converter

• Supports Windows.

Comparison Results:

nagadomi's waifu2x seems to produce the cleanest results with the least amount of noise/artifacts.

• The custom model was trained using nagadomi's waifu2x. The model was trained on images created by Neural-Style, and many of the associated style images.

• Both of the waifu2x-caffe results seem to produce noisier images with more artifacts.

• WL-Amigo's waifu2x-converter seems to produce to cleaner images than waifu2x-caffe, but it's still not as good as nagadomi's waifu2x in terms of artifacts/noise.

Notes:

• Keep in mind that there can be situations where you want more noise/artifacts, or less noise/artifacts.

---