2
$\begingroup$

I'd like to build an application for tracking the position of a given animal (e.g. a cat) in a series of images.

Is there any off-the-shelf API I could use?

Azure has some Vision APIs, but it seems to me they can't be used to get the position of something in an image.

$\endgroup$
3
  • $\begingroup$ When you say position, do you mean something like bounding box coordinates around the animal? The most “off-the-shelf” API I’ve seen is Fast.Ai. They’re always implementing SOTA algorithms, and expanding regularly. They also have lectures for free online outlining their code. It’s a little bit “black box” but might work well for your use case. $\endgroup$
    – Hanzy
    May 31, 2019 at 3:17
  • $\begingroup$ Thx. Just to confirm, this is their site, right: docs.fast.ai. I'm totally fresh to AI, (just want use it on a personal project) so I thought I better double-check. $\endgroup$
    – Joao Silva
    May 31, 2019 at 6:23
  • $\begingroup$ I believe this is their main site: fast.ai. I suppose the one you posted is documentation which will probably be useful. But I think under the MOOC section of the main site you can watch video lectures with examples of their API (and a little theory) $\endgroup$
    – Hanzy
    May 31, 2019 at 12:15

1 Answer 1

2
$\begingroup$

FastAI is the most “out of the box” API for this type of task.

For video examples (and a little theory) check out the MOOC section of their site.

Practical Deep Learning and Cutting Edge Deep Learning are the two sections most relevant to you.

But if you want a working implementation check out this GitHub repo that implements SSD for your purposes. I can’t say how simple the API is but it does what you are seeking (in pure PyTorch).

note: FastAI was originally built on top of PyTorch (although they are expanding out now), so you’ll be using PyTorch and need a rough idea of how to work with tensors. Most of the challenges are already implemented in their api. Installing their library will automatically install PyTorch.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .