> For the complete documentation index, see [llms.txt](https://docs.shootxp.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.shootxp.com/key-features/shooting-ai.md).

# Shooting AI

Upload video footage from a standard camera/device and have our AI model detect: 

* shot attempts
* shot locations
* makes/misses
* contested shots
* movement of the shooter leading up to the shot
* number of players on the court when a shot was attempted

All detected shots are time-stamped and therefore, when imported into the ShootXP app, allow you to quickly find and view the section of video associated with any shot(s).

<div data-full-width="false"><figure><img src="/files/RwCcfIlUBl6uMSiXIqsc" alt=""><figcaption></figcaption></figure></div>

Now, despite the over-enthusiastic claims of some marketers (🤑), AI is far from "magic" and has its share of limitations. Our shooting model is no different and a number of factors can impact how accurately and consistently events are detected from uploaded footage.&#x20;

With this in mind there is a still a need for (and benefit in having) a "human in the loop", so we have put a lot of time and attention into the ShootXP app so that you can:

1. contribute to training a custom model(s) that will work best for the specific camera perspective(s) and venue(s) that you capture footage from, ie your training facility.
2. manually correct (add/remove/edit) any data that the AI model spits out.

&#x20;