# ONNX use cases

ONNX supports integration of trained machine‑learning models into SDK workflows that require portable, framework‑agnostic inference. It is commonly used in media and analytics pipelines where inference is performed as part of video or audio processing. This approach enables deployment across heterogeneous hardware using a standardized model format.

Before you run the use cases, complete the preconditions mentioned in [GStreamer command-line use cases](https://docs.qualcomm.com/doc/80-70029-50/topic/gstreamer-application-use-cases.html).

- [Image classification with ONNX](https://docs.qualcomm.com/doc/80-70029-50/topic/image-classification-with-onnx.html)
The use cases use an ONNX model to perform scene classification within a single‑camera media pipeline, with inference results overlaid on or composed into the output stream.
- [Object detection with ONNX](https://docs.qualcomm.com/doc/80-70029-50/topic/object-detection-with-onnx.html)
The use cases use an ONNX object detection model to identify objects in a scene from a single camera stream. The detected bounding boxes are either overlaid on the video output or composed into the rendered stream and displayed.

Last Published: Apr 02, 2026

[Previous Topic
Image segmentation and encode with Neural Processing SDK](https://docs.qualcomm.com/bundle/publicresource/80-70029-50/topics/single-camera-stream-with-image-segmentation-and-encode-with-deeplabv3-quantized.md) [Next Topic
Image classification with ONNX](https://docs.qualcomm.com/bundle/publicresource/80-70029-50/topics/image-classification-with-onnx.md)