[HN Gopher] Modelplace: AI Model Marketplace by OpenCV
___________________________________________________________________
Modelplace: AI Model Marketplace by OpenCV
Author : philnelson
Score : 113 points
Date : 2021-06-15 12:56 UTC (10 hours ago)
(HTM) web link (modelplace.ai)
(TXT) w3m dump (modelplace.ai)
| ipsum2 wrote:
| 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.
| gieoon wrote:
| I agree, it seems like fine-tuning the massive open source
| models is better than buying small niche models. Curating the
| dataset is also a non-trivial amount of work that has to be
| customised per client. However with further adoption of AI into
| new markets new clients may emerge.
| meragrin_ wrote:
| Are they selling models or making introductions? As you say,
| niche models have small audiences which makes it all the more
| likely the models will require modifications. Discussing those
| modifications provides an opening to adjust the pricing.
| pepe-sad wrote:
| It looks like model-as-a-service. Cool feature, isn't it?
| blululu wrote:
| 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.
| Namidairo wrote:
| Weird, the account registration wouldn't work on my normal IP,
| claiming 'too many requests'
|
| Worked fine on a VPN however (sorry, I know you probably don't
| like this).
|
| 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".
| philnelson wrote:
| Thanks for the report- it looks like we had our anti-spam set a
| little too sensitive. Glad you like the site!
| moritonal wrote:
| 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).
| otoburb wrote:
| >> _Did you really name your company OpenCV? As in.. Open
| Computer Vision, which you know is an incredibly well-known
| brand for a 20-year old library, likely fundamental to most the
| libraries you 're now trying to commission for in your market?_
|
| This looks deliberate because the Chief Scientist & President,
| Gary Bradski[1], created the OpenCV library that you refer
| to.[2]
|
| [1] https://www.oreilly.com/pub/au/3270
|
| [2] https://www.opencv.ai/#About-us
| moritonal wrote:
| Ah, well, that's fair. Egg on my face.
| mendeza wrote:
| A lot of these models are open source, I wonder who would this
| product market towards?
| karxxm wrote:
| The models have to be trained. Trained towards a given task.
| yeldarb wrote:
| And then once you have the weights you have to figure out
| what to do with them. Data science (training a good model)
| and engineering (deploying it as a service to be consumed by
| end users) are pretty distinct tasks.
| pepe-sad wrote:
| Good point.
| numbers_guy wrote:
| Can I ask where one can download the pretrained weights? Is
| there some online repository?
| belval wrote:
| You can look up the torchvision model zoo for pretrained
| weights of most popular architectures.
|
| Tensorflow as an equivalent tool but I can't remember the
| name.
| _joel wrote:
| Some for YOLOv4 https://github.com/AllanYiin/YoloV4
| wolterlw wrote:
| 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.
|
| Having a more standardized description, something like model
| cards [1] (CV example [2]) would be hugely beneficial IMO.
|
| [1] https://arxiv.org/pdf/1810.03993.pdf
|
| [2] https://drive.google.com/file/d/1f39lSzU5Oq-
| j_OXgS67KfN5wNso...
| pepe-sad wrote:
| They looks quite similar. What did you find useful in this card
| and can't get from a model description?
| joshmarlow wrote:
| I hadn't heard of model cards before - thanks for the link!
|
| I wonder how feasible it would be to develop some formal
| ontology to describe the capabilities of models and attach them
| to model cards.
|
| That could open up some interesting applications:
|
| * automated search for a set of models to accomplish some task
|
| * a human usable search engine for finding specific models
|
| EDIT: to complete a sentence
| philnelson wrote:
| It's a good suggestion, thanks.
| rexreed wrote:
| Looks like https://modeldepot.io shut down, which was another
| model-as-a-service marketplace. You can check out
| https://paperswithcode.com/
| version_five wrote:
| I can see value in being able to casually integrate vision models
| into different applications.
|
| 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.
| JSbloater wrote:
| I can see the lawsuits already
| jcims wrote:
| 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.
| philnelson wrote:
| Thanks! This is definitely a feature we want to add in the
| future.
| innerlee wrote:
| We also expored ways of presenting hundreds [1] of AI models in
| a better way. Recently we updated the model zoo so that it is
| arranged according to benchmark datasets [2] and papers [3]. Do
| you think it satisfies the needs?
|
| [1] https://mmpose.readthedocs.io/en/latest/modelzoo.html [2]
| https://mmpose.readthedocs.io/en/latest/topics/body(2d,kpt,i...
| [3]
| https://mmpose.readthedocs.io/en/latest/papers/algorithms.ht...
| wokwokwok wrote:
| How do you stop piracy on something like this?
|
| 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?
|
| ...and I mean, is the problem not utterly fundamental?
|
| When is a model "yours" or sell?
|
| 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?
|
| I see this being problematic to look after...
| Moosdijk wrote:
| Disclaimer: I'm not very familiar with the current state of CV
| and the way models are trained.
|
| What authors could do is train the model on a dataset, but add
| a specific image that it flags as a key.
|
| That way, if you keep the image private, you can check if a
| model is a clone of yours if recognises that specific image.
| belval wrote:
| The value of these could be in private datasets. Model
| architectures are not really patentable and ownership of a
| trained model has always been very muddy water.
|
| That being said, if I am selling a bird classifier chances are
| that to train it I had to combine multiple datasets and someone
| can't easily finetune it to get better results.
| aledalgrande wrote:
| The marketplace should probably be on datasets.
| Moosdijk wrote:
| I'd rather buy a model than a dataset that I have to spend
| money to train a model with.
| aledalgrande wrote:
| You are limited in expanding your dataset and making your
| model better, if you only have the model. And I would
| rather see how clean/balanced the dataset my model was
| built from was.
| wokwokwok wrote:
| They dont need to fine tune it for _better_ results, they
| only need to fine tune it to give _different_ results and
| sell it for $5 less than you are.
| belval wrote:
| I misunderstood your point.
|
| The "attack" that you are describing could probably be
| mitigated by doing activation analysis. Given random
| initialization even with fine-tuning the activation
| distribution will be similar. Same method should be used to
| weed out people just repackaging the ImageNet pretrained
| model from torchvision.
|
| That being said you can still distill a network into
| another with a teacher-student approach and that would be
| undetectable.
| pepe-sad wrote:
| I believe it has a kind of pre-moderation.
| bonoboTP wrote:
| Right, you could steal the model by querying lots of images and
| using teacher-student model distillation.
___________________________________________________________________
(page generated 2021-06-15 23:01 UTC)