# Image segmentation
Source: [https://docs.qualcomm.com/doc/80-70020-50/topic/gst-ai-segmentation.html](https://docs.qualcomm.com/doc/80-70020-50/topic/gst-ai-segmentation.html)
The **gst-ai-segmentation** application allows you to divide an image into
different and meaningful parts or segments and assign a label to each homogenous segment
based on the similarity of the attributes. The application uses Qualcomm Neural Processing
SDK runtime, Qualcomm AI Engine direct runtime, and LiteRT for image
segmentation.
The following figure shows the pipeline, which receives the input from a live camera
feed, file, or an RTSP stream, preprocesses the video data, runs inferences using AI
hardware, and displays the segmented data on the screen.
For information about the plugins used in the pipeline flow, see [Pipeline flow](https://docs.qualcomm.com/doc/80-70020-50/topic/gst-ai-segmentation.html#gst-ai-segmentation__section_xb4_p1s_lbc).
Figure : gst-ai-segmentation pipeline
## Sample model and label files
| Runtime | Model files | Label files |
| --- | --- | --- |
| Qualcomm Neural Processing SDK | deeplabv3_resnet50.dlc | deeplabv3_resnet50.labels |
| LiteRT | deeplabv3_plus_mobilenet_quantized.tflite | deeplabv3_resnet50.labels |
| Qualcomm AI Engine direct | deeplabv3_plus_mobilenet_quantized.bin | deeplabv3_resnet50.labels |
| | | |
| | | |
## Run the application on the target device
The sample application uses the
/etc/configs/config\_segmentation.json file to read the
input parameters.
To create your own config JSON file, use [config_segmentation.json](https://git.codelinaro.org/clo/le/platform/vendor/qcom-opensource/gst-plugins-qti-oss/-/blob/imsdk.lnx.2.0.0.r2-rel/gst-sample-apps/gst-ai-segmentation/config_segmentation.json?ref_type=heads) as a
reference.
1. Ensure that you complete the [Prerequisites](https://docs.qualcomm.com/doc/80-70020-50/topic/download-model-and-label-files.html).
2. Update the config JSON file based on the model, input stream, and other
properties. For more information, see [Config JSON field description](https://docs.qualcomm.com/doc/80-70020-50/topic/gst-ai-segmentation.html#gst-ai-segmentation__section_ict_rdr_32c).
3. Use the following format of the config\_segmentation.json
file:
{
"file-path": "",
"ml-framework": "",
"model": "",
"labels": "",
"constants": "",
"runtime": ""
}Copy to clipboard
For example, run the application using the
DeepLabV3-Plus-MobileNet LiteRT model, DSP runtime,
custom model and label paths, and constants from the video
file:
{
"file-path": "/etc/media/video.mp4",
"ml-framework": "tflite",
"model": "/etc/models/deeplabv3_plus_mobilenet_quantized.tflite",
"labels": "/etc/labels/deeplabv3_resnet50.labels",
"constants": "deeplab,q-offsets=<0.0>,q-scales=<1.0>;",
"runtime": "dsp"
}Copy to clipboard
Note: The values for the
`q-scales` and `q-offsets` constants are
`<1.0>` and `<0.0>`,
respectively.
4. Run the gst-ai-segmentation application:
gst-ai-segmentation --config-file=/etc/configs/config_segmentation.jsonCopy to clipboard
5. To display the available help options, run the following command in the SSH
shell:
gst-ai-segmentation -hCopy to clipboard
6. To stop the use case, use CTRL +
C.
## Expected output
The segmented data is displayed on the local display.
Figure : Expected output for gst-ai-segmentation application
## Pipeline flow
The following table lists the plugins used in the image segmentation
pipeline:
| Plugin | Description |
| --- | --- |
| Camera source:[qtiqmmfsrc](https://docs.qualcomm.com/doc/80-70020-50/topic/qtiqmmfsrc.html) |
Captures the live stream from camera
Uses tee to split the stream for inferencing
|
| File source: filesrc |
Captures the video stream using filesrc, followed by qtdemux, which demultiplexes the stream
Uses tee to split the stream for inferencing
|
| RTSP source: rtspsrc |
Captures the RTSP stream using rtspsrc, followed by rtph264depay for video extraction
Uses tee to split the stream for inferencing
|
| h264parse | Parses the H.264 video |
| [v4l2h264dec](https://docs.qualcomm.com/doc/80-70020-50/topic/v4l2h264dec.html) | Decodes the video |
| [qtimlvconverter](https://docs.qualcomm.com/doc/80-70020-50/topic/qtimlvconverter.html) |
Receives the video stream on its sink pad
Performs the following preprocessing on the stream data. This preprocessing is done when the model expects floating-point values as input
Color conversion
Scaling (up or down)
Normalization
Converts the preprocessed video stream to a tensor stream on its source pad
The tensor stream is used for inferencing in the later stages of the pipeline. |
| Inferencing plugins:
After the inference runtime receives the tensor stream on its sink pad, it runs the inference
Produces a tensor stream with the inference results on its source pad
|
| [qtimlvsegmentation](https://docs.qualcomm.com/doc/80-70020-50/topic/qtimlvsegmentation.html) | Converts the inference tensors that it receives on its sink pad into video formats that the multimedia plugins use for further processing |
| [qtivcomposer](https://docs.qualcomm.com/doc/80-70020-50/topic/qtivcomposer.html) |
Composes frames by combining content from its sink pads
Pushes the GStreamer buffers containing these composed frames to its source pad
Forwards the video stream received on its sink pad to Weston
Weston renders the video stream on a local display
|
## Config JSON field description
The different parameters available to configure the JSON file and run the use case
are as follows:
Table : Field description–config_segmentation.json file
| Field | Values/description |
| :--- | :--- |
| **ml-framework** | Enable and use one of the following models:
snpe: Qualcomm Neural Processing SDK
tflite: LiteRT
qnn: Qualcomm AI Engine direct
|
| **runtime** | Enable and use one of the following runtimes:
cpu
gpu
dsp
|
| **Input source** | Enable and use one of the following input sources:
camera: Primary (0) or secondary (1).
file-path: The directory path to the video file.
rtsp-ip-port: The address of the RTSP stream: rtsp://<ip>:<port>/<stream>
|
## Known issue
An FPS drop is observed when using the Qualcomm Neural Processing Engine SDK
models.
## Related information
- [Image segmentation and display with LiteRT](https://docs.qualcomm.com/doc/80-70020-50/topic/single-camera-stream-with-image-segmentation-and-display.html)
- [Image segmentation and encode with LiteRT](https://docs.qualcomm.com/doc/80-70020-50/topic/single-camera-stream-with-image-segmentation-and-encode.html)
- [Image segmentation and display with Neural Processing SDK](https://docs.qualcomm.com/doc/80-70020-50/topic/single-camera-stream-with-image-segmentation-and-display-with-deeplabv3-quantized.html)
- [Image segmentation and encode with Neural Processing SDK](https://docs.qualcomm.com/doc/80-70020-50/topic/single-camera-stream-with-image-segmentation-and-encode-with-deeplabv3-quantized.html)
**Parent Topic:** [Run AI/ML sample applications](https://docs.qualcomm.com/doc/80-70020-50/topic/ai-ml-sample-applications.html)
Last Published: Jan 30, 2026
[Previous Topic
Pose detection](https://docs.qualcomm.com/bundle/publicresource/80-70020-50/topics/gst-ai-pose-detection.md) [Next Topic
Image segmentation using Python with container](https://docs.qualcomm.com/bundle/publicresource/80-70020-50/topics/image-segmentation-using-python.md)