Coloring Old images



Taking an old B&W  image and coloring it has become a thing in recent years. When done well it can add life to a image and bring new insights. However generally it takes a lot of work in photoshop and a trained technician. However even done well, it can have issues because often the colorer has to guess on the color of clothes etc

I was interested to learn therefore of work based on machine learning to do the same thing. The idea is to train a algorithm based on a set of images and then use the result to color unrelated images.

I was skeptical initially, but the results looked good so I thought I would have a closer look. I wanted to see if I could run the process myself. The code can be found here

https://github.com/jantic/DeOldify

and instructions are here

https://github.com/jantic/DeOldify/blob/master/README.md

 However it looked like a lot of effort to set it up, but fortunately there was an easier way

If you go to this link

https://colab.research.google.com/github/jantic/DeOldify/blob/master/ImageColorizerColab.ipynb

You can run a virtual machine totally in your browser. If you click it you will see  a page like this

All you need to do is click the boxes in the right hand column. Each time it will do some part of the setup is carried out. Wait till is complete then do the next one

Finally you will reach this section



This is where you do your work. First you need to provide an image.

At this point I have not found a way to load an image from disk so you need to upload your images to the internet. An important fact however is that the URL of the image must end in .jpg. I use imgur.com as my repository which is free. Upload your image and copy the URL into the source url box and then hit the start button under Colorize. After a few minutes it should present you with a screen containing a full resoloution coloured image, a smaller version and the original B&W

I have not found a way of saving the image directly, so instead right click on the full res version and hit save as. You can then save it as a .png file

If you have problems, try reducing the size of the original image.

The results


The results vary depending on the source of the original.  The better the scan, generally the better the image, however the compare well with the hand created version. Plus its free

You can also play with the render factor controls.














Comments

Post a Comment

Popular posts from this blog

A Bridgen too far

Image
  The Life and Death of Colonel Bridgen For the last 14 years, I have been represented in Parliament by Mr Andrew Bridgen MP. For the vast majority of my life I, like so many, have been largely apolitical. You cast your vote in the various elections (The next general one is only a few weeks away), and after that, apart from general grumbles, you ignore the result. As the late Douglas Adams put it, you vote for the lizard, because if you don't, the wrong lizard gets elected. However, being part of Andrew's constituency has changed the way I look at politics. Let me explain.. In the beginning... We moved into NW Leicestershire in about 1990 which at the time was were represented by David Taylor MP.  David was a typical back bench MP. A local guy, working hard to represent the area, and his only ambition was to do the best job he could. We would quite often bump into him in the village, either hand-delivering leaflets, or just having a coffee in a church café. Whatever his p
Read more

Getting a NVME card ready

Image
    Ok, the next challenge. Installing the software Now there are an easy way to do this, and hard ways. This is the hard way. Why not take the easy way?  Well, you may have noticed, people like me are masochists looking for a hit. This is what keeps us going. Also, this is the learning curve. It is all very good, to have tools automate everything, but when the poo hits the rotating object, it is good to know why things work We will break down this into 3 stages 1. Install the Pi OS 2.  Get the M.2 NVME board recognized 3. Install the Pi OS onto it To do this, I largely followed To do this, I followed this tutorial . However, there were some issues, so we will work through them Step 1.Install Pi OS   Note: you must do this step before you build your Argon box, because you cannot access SD card slot from outside the case    Note:  use a reasonable size card (e.g 32 GB) since we need to load an OS image onto it The 1st step is to install our initial Pi OS. To start with, we mu
Read more

A new case study

Image
  I have played around with Linux for about 20 years, and I am constantly amazed by its ability to make simple things hard Why? Well, I tell myself it's money, but in truth it is more about the achievement of making things. Maybe it is a holdover from the days men went to kill woolly mammoths, but that feeling of overcoming obstacles feels great, even if the end product is not as slick as a bought product. Unfortunately (or maybe fortunately), installing anything with Linux will have issues. Some things procedures have been smoothed out over the years, but the more you dive into it, the more issues you will hit. It's not as if, however, there is no advice. In fact, it is the opposite. There is far too much, but many of them are badly written assuming some critical knowledge or out of date dealing with legacy software etc. However, as Mission Impossible says, "Your mission, should you choose to accept it..". Start the theme music now The challenge For example, a few y
Read more