I would like to request that Tulip Vision be made usable with cameras other than RGB‑D cameras. Specifically, I hope that Tulip Vision AI can operate with cameras such as iPhone and iPad cameras, IP cameras, and web cameras, including those that do not provide depth information.
From my understanding, the Depth (D) information is mainly used to determine the distance to objects recognized by the AI. If it were possible to define a detection region (detection area) instead of relying on distance information, use cases such as detecting whether a person is standing in front of a machine tool could still be fully satisfied.
Therefore, I would like you to consider supporting cameras that do not require depth information. In addition, while Tulip Vision is currently described as operating on Windows 11, I would also like to request future support for iOS devices (iPhone and iPad).
Regarding iPhone/iPad cameras specifically, those are not currently supported, and Vision does require Windows 10/11. Are you working solely with iOS devices for your operations, or do you have Windows devices available as well? If you have Windows machines on the floor, you may already have more options than you realized!
We are assuming a future scenario in which custom models are run on iOS devices.
For example, we are considering operations where AI automatically counts the number of resin pallets or metal carts based on photos taken with a smartphone.
In this context, we would like to ask whether the main limitation would be related to the Tulip application itself, or whether it would be constrained by the performance capabilities of the iPhone.
Thank you again for your response, and we look forward to continuing the discussion.
Today, Tulip Vision runs on Windows 10/11, so for your pallet and cart counting use case, you would set it up using a Windows machine with a supported camera. That said, we have other AI solutions that could help with such use cases, including AI agents. Have you had a chance to experiment with these?