Hello! I’m a long time developer but new to AI-based image processing. The end goal is to process images from cameras and alert when deer (and eventually other wildlife) is detected.
The first step is finding a decent model that can (say) detect deer vs. birds vs. other animals, then running that somewhere. The default The CameraTraps model here allows detecting “animal” vs. “person” vs. “vehicle”:
Would I need to train it further to differentiate between types of animals, or am I missing something with the default model? Or a more general question, how can you see what a frozen model is set up to detect? (I just learned what a frozen model was yesterday)
Appreciate any pointers or if there’s another sub that would be more suited to getting this project setup, happy to post there instead 🙂