4
12 Comments

Enable creators to see who's stealing their images

Hi everyone,

I know a lot of independent creatives that are for example illustrators or photographers. The way they make money is by making images for clients and selling them with a certain "rights of use". However, these are hard to negotiate and often violated.

My idea was to make a web app where these creatives can upload their work and then it would scan the web to see if these appear on websites and notify the creator. The creator can then either approve or disapprove the content. If they disapprove they can reach out to the creator of the website. Later we could even provide legal services or create a portal in which creatives can safely distribute their images to clients with contracts etc. But this first monitor tool would already be valuable enough to get people paying for it. I've pitched it to multiple illustrators and they all really liked the idea and would pay for it (interesting side note is that it should be fixed monthly fee, not pay per image as some of them produce way more than others).

MVP Tech Execution

  • We could use a wrapper like this to take care of subscriptions: https://www.memberstack.io
  • Reverse image lookup is the hardest challenge. Google does a great job but doesn't allow bots. Others like TinEye don't seem to be good enough to spot all images. Maybe there are other ways to do this? I think this core tech should feel magical. If not, I think tech is not mature enough for this product to be a success.
  • There's a some APIs that allow you to get associated e-mail based on a URL to reach out.

Would love to know what you think of this? and if you have ideas on execution? If you want to partner up, let me know as well.

on January 29, 2020
  1. 2

    For most independent creatives, this kind of service is more like a punch to the face.

    You show them the 10,000+ shady places on the internet where their content has turned up, but they are helpless to do anything about it due to their own past actions. FYI, CTO of ImageRights.com here, so basically I've been running such a business for many years. But of course, I'll post my personal opinion here.

    My impression is that most independent content creators sell themselves below value (e.g. $5 stock photos) and are lazy when it comes to registering with their local copyright office. As a result, even if they are 100% in the right, it is challenging to find an attorney in the EU or US to take on their claim, due to the low value and the lack of registration.

    Because of that, the customers that derive value from such a service are usually professional freelancers and agencies. And even they derive most of the value not from us showing them where their images turned up, but from us having a legal network to help them with doing anything about it.

    You'd be surprised at how many large and supposedly respectable companies will just steal someone's content and then outright refuse to purchase a license for it when you contact them. So you definitely need attorneys to file court cases, if you plan to generate a financial benefit for your customers.

    And when you get there, there's actually established competition here:
    ImageRights, PixRay, PicRights, imageDEFENDERS, COPYTRACK, Pixsy, Higbee, Binded

    BTW, the angle of using Blockchain or Crypto to "secure" your images has been beaten to death already. It doesn't work, but that did not stop people from running shady ICOs, so now potential customers avoid Blockchains like the plague.

    Oh and lastly, the companies stealing content at a larger scale tend to be competent enough to block Google / Bing from indexing the stolen images... Last time I checked, roughly 80% of the image uses that our customers sue over did not show up on Google. So you'll need your own crawling infrastructure if you intend to generate halfway complete reports. And that means the price to enter the market becomes quite high, as you'll need lots of servers with lots of RAM and proprietary algorithms.

    In my opinion, TinEye is actually doing quite well, but they are very expensive. LTU was cheaper but a lot worse quality. We use our own algorithms developed in-house since 2010. The general approach of "take each web image and compare with each customer image" is not that difficult to get to work, but once you reach thousands or millions of customer images, it will be way too slow. So what you're left with is building your own distributed database system, which I think was a really cool project :)

  2. 1

    As an illustrator/photograph, why would I pay for this service, if I can just manually check on 'search by image' on Google, and locate all the websites that have used my illustration/photo ?

    1. 1

      Convenience is key.
      You can totally do that, but that's quite some work that you could also spend on making money. Especially for people that have made a lot of work over the years, keeping track of it all is very labor intensive.

      1. 1

        Well, I don't agree actually...
        How much time would it save me ? 5-10 minutes per week ?
        Also, if I upload my artwork to your platform to get the results, it would be nearly the same as uploading my artwork to 'search by image' on Google.
        It is nice idea, but I'm trying to understand the uniqueness of the solution.
        How much faster will it be ? What can I do with the saved time, by using this solution ? Would I make x5 profit with my saved time if I use your solution, instead of doing it manually ?

        1. 1

          Thanks for the feedback. This tool would proactively send you an e-mail when your work turns up somewhere else. That is convenient:

          Also, if I upload my artwork to your platform to get the results, it would be nearly the same as uploading my artwork to 'search by image' on Google.

          True, but if your work is on 20+ websites and you have 100+ original assets, that is quite labour intensive. That's a 100 searches, and a few pages you have to scan per search, I don't think many people will do that voluntarily.

          After finding your images on places where they don't belong you would either be able to 1. sell a license to these people or 2. proceed with legal steps.

          1. 2

            "This tool would proactively send you an e-mail when your work turns up somewhere else" - this is the selling point here!

  3. 1

    I could see a service working for website owners where you enter a URL and it crawls your website and cross checks your images. This service would be for free for the first run through and you could offer a monthly subscription scan for a fee

    1. 1

      Wow, thanks for sharing that idea. Really like this approach, because it's even more automatic and magical than adding the assets manually.

  4. 1

    Awesome idea @samuelbeek .
    I tried using Google and Bing with a simple modern SaaS illustration and the results are not that great for similar images: https://bit.ly/36CVE8r

    Maybe the tech is still not there but with the advance of ML it will definitely be possible soon.

    1. 1

      Yeah that's the problem - if it's not right 90% of the time, I don't think it will add value.

    1. 1

      Thanks - this is very useful :) I will have to look at the performance, but the price seems reasonable. $7 for a 1000 requests could be doable for this project.