Styjun

The trick I use is loading the image in Google Docs. It's time consuming, but it works. Here are the steps to use:

1. Edit the post so you can see the source. Grab the image url (either inline or at the bottom of the post.


2. Open a Google Document, preferably a Word Document.


3. In the "Insert" menu, choose "Image".
   


4. On the left menu, choose "By Url", then paste the url in the box. If all goes well, you will see the image. You can insert it into the document to see it larger.
   

Bewarned about links that don't go to i.stack.imgur. If the link goes to a page that contains the image, then this trick won't work. Then you just have to wait until you get home.

The reason why this works is because it's Google's servers that are downloading the picture, which can get to the sites that are blocked for you.

Another proxy is using duckduckgo image results page,

The img link: https://i.imgur.com/f01Ert6.jpg

Becomes like this: https://proxy.duckduckgo.com/iu/?u=https://i.imgur.com/f01Ert6.jpg

So inspired by James' answer about using Greasemonkey, I wondered how tough it'd be to hack up a quick JavaScript snippet that'd swap out image sources.

The below code will swap out all images whose sources start with i.stack.imgur with their counterparts from archive.org. To be clear, this puts the images on the page you're viewing, which is ideal, I think.

$("img").each(function(ignore, e) 
 var $e = $(e);
 if ($e.attr("src").startsWith("http://i.stack.imgur")) 
 $e.attr("src", "http://web.archive.org/web/" + $e.attr("src"));
 
);

To run this easily, you'll need to access a developer console for your browser of choice (often brought up with F12 if on Windows). I'm using Firebug, but I bet Firefox's normal, built-in tools work too.

_{(You could also use Greasemonkey, though this is easy enough I'm not going to bother. It looks like recent versions of Greasemonkey have some trouble, and it's not a big deal to bring up dev tools. Still, Greasemonkey 2.3.1 seems stable, and it'd be easier to use than having this code in a txt file somewhere to paste over and over.)}

Note: Originally I was going to try some way of using http://images.google.com, but that's blocked for me right now too! Archive.org is not, though my guess would be that fewer images are stored there than images.google, and I wonder how quickly answers are indexed. You'll still miss some images, especially recent ones, using this technique.

This did work for the image-intensive answer I'm currently viewing. Wish I'd done this months ago. Stupid. ;^) Guess I should go contribute now.

If you're running a browser that supports Greasemonkey and you're able to get to an open proxy service, you should be able to write a page script that munges all the imgur URLs.

I found someone on reddit with a similar problem that posted a script that might work.

You can use FoxyProxy (also exists for Chrome) and do a SSH port forwarding to your machine at home ;)

To create the tunnel:

ssh -D 8080 you@yourserver.com

Then you use FoxyProxy to connect FireFox/Chrome to your local port 8080 (if you have no admin rights, use a Port > 10'000).

If port 22 (SSH) is blocked, you can set the SSH daemon at home to listen at the SSL port, that almost always works, at least for me ;)

You can also tell Firefox to use the proxy for DNS-resolution already ;)

For Windows, you need PUTTY for the SSH tunnel:
http://www.hostdime.com/resources/browsing-internet-ssh-tunnel-windows/
or you can use the ssh in git-bash in git-scm. Works fantastic.
With SSL-Port: ssh -D 10001 username@yourserver.com -p 443

With git-scm, you can even use RSA private-public keys, especially when you don't have admin rights, and going through putty-gui is just too slow.

To generate a RSA-key (4096 bit) for the ssh-daemon, execute

ssh-keygen -t rsa -b 4096
cat ~/.ssh/id_rsa.pub >> authorized_keys

you need to generate the key on the client, then take the id_rsa.pub in the ~/.ssh folder and echo the text in id_rsa.pub in ~/.ssh/authorized_keys on the server for the user you want to use.

That's much safer than using a PUBLIC proxy that can't be trusted (those cursed moments when you forget to switch the proxy off), and you can switch it on/off in an instant.

Also, if you connect to the SSH server, use the IP instead of the servername, that way DNS-blocking will not be able to stop you.


If you have no admin rights, use PortableApps (for both Firefox and PuTTY, Chrome doesn't need admin rights for installation).

All the network admin will see is a SSL connection to your home server IP.

That's much safer anyway.

No more monitoring of your browsing activity, no more blocked sites, no more traces.

Everything is encrypted.

NOTE: This is somewhat outdated. It still works fine as of 5/1/2018, but I've written another userscript that seems to work a bit better.

I've created a new tampermonkey script specifically for this purpose called Image Proxier.

This will replace all links from ["https://i.stack.imgur.com/", "https://i.imgur.com/"] with links from http://web.archive.org/web/ (by default, this can be changed relatively easy).

This userscript works for images both in questions/answers, and also comments. In comments it will convert links to images into a thumbnail and collate them all together at the top of the comment. Here is an example of comments with 1 or 2 images:

You can then click on the image to see a bigger version of it:

On images inside questions, it will similarly collate all of the images together, but only ones near eachother in paragraphs. If multiple images exist in a paragraph, they will all be collated together right above where they originally had been.

NOTE: Images posted may not immediately be picked up by archive.org, they may take some time to actually display the picture.

I've created another Userscript that accomplishes this, and I believe it does a better job than my other answer because it avoids the issue of newer images not loading by actually sending the picture to a proxy service which returns a working image.

I've made a userscript that I call Image Wizard. Image Wizard will take links and images from given domains with a real proxied version of the image (currently from bypass123.com, but this is easily changeable.)

This will make all images (from imgur.com and facebook.com by default) inside content posts on Stack Overflow look like this:

And links in comments will be changed to [Image Wizard]:

Any link/image that is converted by Image Wizard when clicked will open in a new tab.

The project is hosted on GitHub.

Image Wizard can be installed through GreasyFork.

Update 5/2/18 - Added an automated method of detecting when images are added to DOM instead of running the function every x seconds. Method - new version v1.5.

If you have a home computer, it wouldn't be too hard to whip up a node.js program that'd let you run a proxy.

In fact...

req.pipe(request(siteURL)).pipe(resp)

...that's a one line proxy right there. You could simply bounce the url through to the server with a command line, or maybe write a simple webpage that'd go ahead and display the image for you.