Objekte in Bildern per Prompt mit Gemini 2.0 erkennen

## 2. Use Prompt-Based Object Detection Read more about the HTTP node We've had generalised object detection before (see my other template using ResNet) but being able to prompt for what you're looking for is a very exciting proposition! Not only could this reduce the effort in post-detection filtering but also introduce contextual use-cases such as searching by "emotion", "locality", "anomolies" and many more! I found the the output json schema of { "box_2d": { "type": "array", ... } } works best for Gemini to return coordinates.