[Bug] When I use mmdeploy_runtime for object detection inference

[tankchenggeng]

Prerequisite

• I have searched Issues and Discussions but cannot get the expected help.
• The bug has not been fixed in the latest version(https://github.com/open-mmlab/mmpose).

Environment

Not Run

Reproduces the problem - code sample

from mmdeploy_runtime import Detector, PoseDetector

detector = Detector(model_path=det_model_path, device_name=device_name)
bboxes, labels, _ = detector(image)

filter detections

keep = np.logical_and(labels == 0, bboxes[..., 4] > bbox_thr)
bboxes = bboxes[keep, :4]

Reproduces the problem - command or script

just code demo

Reproduces the problem - error message

No

Additional information

I have two questions that I don't know how to do:

1. I want to use image mask apply to detector, how should I do?
   2.I want to detect just for person, rather than filter with label==0