Dear use-IP Forum,
We’d like to briefly introduce the benefits of the CAM-AI software, which could be a great solution for those of you who are not fully satisfied with existing options.
CAM-AI allows you to train cameras with artificial intelligence to recognize specific objects. Through continuous learning and image classification, CAM-AI software can not only recognize pre-trained categories such as human, car, truck, bicycle, dog, and cat but also learn to identify new objects. This capability enables the triggering of targeted alarms, whether in the form of email notifications or integration with smart home devices like Philips Hue or Shelly. This significantly reduces false alarms, such as those caused by insects in front of the camera lens.
You can find more information on our website at Artifical Intelligence for Cameras - No more false alarms, along with two demo streams to give you an initial impression: CAM-AI client.
What’s particularly exciting is that we will soon be expanding our range of pre-trained categories to include many new objects and animals. Additionally, it will be possible to define and train custom categories.
Our software is available as an open-source project on GitHub and is free to use for our early users. You have the option to either use our web browser app with a European server location or install the software on your own server. The hardware for your own server is scalable. While 2-3 cameras can already run on a Raspberry Pi (4+), for a larger setup (like connecting NVRs), we recommend a more powerful Debian PC, as detailed in our instructions.
We hope that our CAM-AI software will be helpful to many of you! If you have any questions, feel free to reach out.
Best regards from Germany,
Sebastian
We’d like to briefly introduce the benefits of the CAM-AI software, which could be a great solution for those of you who are not fully satisfied with existing options.
CAM-AI allows you to train cameras with artificial intelligence to recognize specific objects. Through continuous learning and image classification, CAM-AI software can not only recognize pre-trained categories such as human, car, truck, bicycle, dog, and cat but also learn to identify new objects. This capability enables the triggering of targeted alarms, whether in the form of email notifications or integration with smart home devices like Philips Hue or Shelly. This significantly reduces false alarms, such as those caused by insects in front of the camera lens.
You can find more information on our website at Artifical Intelligence for Cameras - No more false alarms, along with two demo streams to give you an initial impression: CAM-AI client.
What’s particularly exciting is that we will soon be expanding our range of pre-trained categories to include many new objects and animals. Additionally, it will be possible to define and train custom categories.
Our software is available as an open-source project on GitHub and is free to use for our early users. You have the option to either use our web browser app with a European server location or install the software on your own server. The hardware for your own server is scalable. While 2-3 cameras can already run on a Raspberry Pi (4+), for a larger setup (like connecting NVRs), we recommend a more powerful Debian PC, as detailed in our instructions.
We hope that our CAM-AI software will be helpful to many of you! If you have any questions, feel free to reach out.
Best regards from Germany,
Sebastian
