Tool Agentic Od

Performs agentic object detection on images to identify and locate objects based on text prompts.

Args: data (FormData): The form data containing the following fields: - prompts (List[str]): List containing exactly one text prompt describing what objects to detect. Must contain between 1 and 1 elements. - image (UploadFile): The image file to analyze for object detection. Required field, cannot be None. - video (UploadFile, optional): Video file input (not supported for agentic OD). Will return an error if provided. baseten_inference_sender: Dependency Injection for sending inference requests to the Baseten model server for agentic object detection.

Returns: ODResponse | JSONResponse Success: ODResponse containing detected objects with bounding boxes and labels. Error: JSONResponse with error details and appropriate status code.