# AI sample applications

The QIR SDK provides the following AI sample applications.

- [Detect hands with `sample_hand_detection`](https://docs.qualcomm.com/doc/80-90441-2/topic/detect-hands-with-sample_hand_detection.html)
`sample_hand_detection` uses the Python-based hand recognition ROS node to detect hand and hand movement. It uses [Qualcomm AI Engine Direct SDK (QNN)](https://www.qualcomm.com/developer/software/neural-processing-sdk-for-ai) for model inference.
- [Estimate image depth values with `sample_depth_estimation`](https://docs.qualcomm.com/doc/80-90441-2/topic/estimate-image-depth-values-with-sample_depth_estimation.html)
This sample application allows you to input an RGB image named `input_image.jpg` or subscribe to the ROS topic `/cam0_stream1` from `qrb_ro_camera`. It then uses QNN to perform model inference and publishes the result as the `/depth_map` ROS topic containing per-pixel depth values.
- [Detect faces with `sample_face_detection`](https://docs.qualcomm.com/doc/80-90441-2/topic/detect-faces-with-sample_face_detection.html)
The `sample_face_detection` sample application is a Python-based face detection ROS node that uses QNN for model inference.
- [Estimate human poses with `sample_hrnet_pose_estimation`](https://docs.qualcomm.com/doc/80-90441-2/topic/estimate-human-poses-with-sample_hrnet_pose_estimation.html)
The `sample_hrnet_pose_estimation` sample application provides high-precision human pose estimation capabilities.
- [Classify images with `sample_resnet101`](https://docs.qualcomm.com/doc/80-90441-2/topic/classify-images-with-sample_resnet101.html)
The `sample_resnet101` is a Python-based ROS node that performs image classification using QNN-based inference.
- [Detect objects with `sample_object_detection`](https://docs.qualcomm.com/doc/80-90441-2/topic/detect-objects-with-sample_object_detection.html)
The `sample_object_detection` is a Python launch file utilizing QNN for model inference. It demonstrates camera data streaming, AI-based inference, and real-time visualization of object detection results.
- [Segment objects with `sample_object_segmentation`](https://docs.qualcomm.com/doc/80-90441-2/topic/segment-objects-with-sample_object_segmentation.html)
The `sample_object_segmentation` is a Python launch file utilizing QNN for model inference. It demonstrates camera data streaming, AI-based inference, and real-time visualization of object segmentation results.

Last Published: Sep 07, 2026

[Previous Topic
Enable people tracking in the simulator with simulation\_follow\_me](https://docs.qualcomm.com/bundle/publicresource/80-90441-2/topics/enable-people-tracking-in-the-simulator-with-simulation_follow_me.md) [Next Topic
Detect hands with sample\_hand\_detection](https://docs.qualcomm.com/bundle/publicresource/80-90441-2/topics/detect-hands-with-sample_hand_detection.md)