I can see value in being able to casually integrate vision models into different applications.<p>I would urge everyone involved (OpenCV and people using the models) to not use this in any safety related application. The example given on the site is person detection for construction site safety. If you're going to use ML for something like this, hire someone with experience using it for such applications, that can understand the operating condition and failure modes. Your worksite is not safe if you just plugged in an online person detection model to a video feed.
If this becomes something like huggingface for Computer Vision - awesome.
For now, it's a shame model descriptions seem to often be lifted straight from README.<p>Having a more standardized description, something like model cards [1] (CV example [2]) would be hugely beneficial IMO.<p>[1] <a href="https://arxiv.org/pdf/1810.03993.pdf" rel="nofollow">https://arxiv.org/pdf/1810.03993.pdf</a><p>[2] <a href="https://drive.google.com/file/d/1f39lSzU5Oq-j_OXgS67KfN5wNsoeAZ4V/preview" rel="nofollow">https://drive.google.com/file/d/1f39lSzU5Oq-j_OXgS67KfN5wNso...</a>
I don't think this will be a successful business model. Niche AI models cost a lot of money to gather the right dataset and train, but they would probably only sell to <100 companies/individuals. So pricing a model at $50-100 isn't profitable. If you're selling general CV models, you have to compete with the plethora of open source ones, including ones that Google and Facebook release for free, which are trained on huge datasets and using a lot of compute resources.
How do you stop piracy on something like this?<p>Surely I could take a model, refine it slightly to get distinct weights (ie. not a flat clone) and unload it as my own novel variant?<p>...and I mean, is the problem not utterly fundamental?<p>When is a model “yours” or sell?<p>When you trained it? When you’re not using transfer learning? When you wrote the paper? When you own the training data? When you can recreate it from scratch?<p>I see this being problematic to look after...
Cool idea. I keep looking for a search/filter/facet function to leverage the tagging and classification they've done with the models, but I can't find one.
Weird, the account registration wouldn't work on my normal IP, claiming 'too many requests'<p>Worked fine on a VPN however (sorry, I know you probably don't like this).<p>After the account registration though, I did quite like trying out a couple models out on a couple frames of footage to see if the models would "work".
Looks like <a href="https://modeldepot.io" rel="nofollow">https://modeldepot.io</a> shut down, which was another model-as-a-service marketplace. You can check out <a href="https://paperswithcode.com/" rel="nofollow">https://paperswithcode.com/</a>
Interesting concept. Would be great to have more technical details about how a model performs. Having a sense of the size, accuracy, platforms supported, and other simple benchmarks would be really helpful to evaluate a model before making a purchase.
This seems like an attempt to cash in and steal the branding behind the library, OpenCV. (I realise now it's the actual authors of the libary.. I've become too jaded here).