# Image segmentation 

Source: [https://docs.qualcomm.com/doc/80-70015-50/topic/gst-ai-segmentation.html](https://docs.qualcomm.com/doc/80-70015-50/topic/gst-ai-segmentation.html)

The **gst-ai-segmentation** application enables 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 shows how to use Qualcomm Neural
        Processing SDK runtime and TFLite runtime for image segmentation.

The figure shows the pipeline, which receives the input from a live camera feed, file, or
            an RTSP stream, preprocesses the video data, runs inference using AI hardware, and
            displays the segmented data on the screen.

For information on the plugins used in the pipeline flow, see [Pipeline flow](https://docs.qualcomm.com/doc/80-70015-50/topic/gst-ai-segmentation.html#gst-ai-segmentation__section_xb4_p1s_lbc).

Figure : gst-ai-segmentation pipeline
            
            ![](data:image/jpeg;base64,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)

## Prerequisites

- To run the application, push the model and label files to the device. For
                    information on downloading the models, see the following:
    - [Download model and label files for Qualcomm Neural Processing SDK](https://docs.qualcomm.com/doc/80-70015-50/topic/ai-ml-sample-applications.html#ai-ml-sample-applications__section_chl_dgz_scc)
    - [Download model and label files for TFLite from AI Hub](https://docs.qualcomm.com/doc/80-70015-50/topic/ai-ml-sample-applications.html#ai-ml-sample-applications__section_fsl_lgz_scc)

    The application supports both the Qualcomm Neural Processing SDK and
                        TFLite models.
- To access your host device, enable SSH. For instructions, see [Use SSH](https://docs.qualcomm.com/bundle/publicresource/topics/80-70015-254/how_to.html#use-ssh).
- Enter the SSH shell and run the use cases:

        ssh root@<ip-addr of the target device>Copy to clipboard
- Enable the
                    display:

        export XDG_RUNTIME_DIR=/dev/socket/weston && export WAYLAND_DISPLAY=wayland-1Copy to clipboard
- Push the files from the host
                    machine:

        scp <filename> root@<IP address of target device>:/opt/Copy to clipboard

Note: For QCS9075, the camera use cases are not supported. Use
                    file or RTSP as input sources.

## Use cases

Note: The following commands provide the default model and label
                paths. If you have a different folder structure, replace the default paths in the
                command-line parameters.

- Run the deeplabv3\_resnet50 model using Qualcomm Neural
                    Processing SDK runtime from the primary
                    camera:

        gst-ai-segmentation --ml-framework=1 --model=/opt/deeplabv3_resnet50.dlc --labels=/opt/deeplabv3_resnet50.labelsCopy to clipboard
- Run the deeplabv3\_mobilenet model using TFLite runtime from
                    the secondary
                    camera:

        gst-ai-segmentation --ml-framework=2 --model=/opt/deeplabv3_plus_mobilenet_quantized.tflite --labels=/opt/deeplabv3_resnet50.labelsCopy to clipboard
- Run the deeplabv3\_resnet50 model using Qualcomm Neural
                    Processing SDK runtime with user model, label file, and input from the RTSP
                    stream:

        gst-ai-segmentation --rtsp-ip-port=rtsp://192.168.1.110:8554/30fps.mkv --ml-framework=1 --model=/opt/deeplabv3_resnet50.dlc --labels=/opt/deeplabv3_resnet50.labelsCopy to clipboard
- Run models using TFLite with different runtimes and input from the primary
                    camera:

        gst-ai-segmentation --camera=0 --ml-framework=2 --use_cpuCopy to clipboard

        gst-ai-segmentation --camera=0 --ml-framework=2 --use_gpuCopy to clipboard

        gst-ai-segmentation --camera=0 --ml-framework=2 --use_dspCopy to clipboard

Display the available help options:

    gst-ai-segmentation -hCopy to clipboard

To stop the use case, press CTRL + C.

## Expected output

The segmented data is displayed on the local display.

Figure : Expected output for gst-ai-segmentation application
                
                ![](data:image/png;base64,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)

## Pipeline flow

The table lists the plugins used in the image segmentation pipeline:| Plugin | Description |
| --- | --- |
| Camera source:[qtiqmmfsrc](https://docs.qualcomm.com/doc/80-70015-50/topic/qtiqmmfsrc.html) | <ul class="ul" id="gst-ai-segmentation__ul_zyl_gj1_mcc"><br>                                    <li class="li">Captures the live stream from camera.</li><br><br>                                    <li class="li">Uses tee to split the stream for inferencing.</li><br><br>                                </ul> |
| File source: filesrc | <ul class="ul" id="gst-ai-segmentation__ul_z1z_x4f_w1c"><br>                                    <li class="li">Captures the video stream using filesrc, followed by<br>                                        qtdemux, which demultiplexes the stream.</li><br><br>                                    <li class="li">Uses tee to split the stream for inferencing.</li><br><br>                                </ul> |
| RTSP source: rtspsrc | <ul class="ul" id="gst-ai-segmentation__ul_vsj_2r4_tbc"><br>                                    <li class="li">Captures the RTSP stream using rtspsrc, followed by<br>                                        rtph264depay for video extraction.</li><br><br>                                    <li class="li">Uses tee to split the stream for inferencing.</li><br><br>                                </ul> |
| h264parse | Parses the H.264 video. |
| [v4l2h264dec](https://docs.qualcomm.com/doc/80-70015-50/topic/v4l2h264dec.html) | Decodes the video. |
| [qtimlvconverter](https://docs.qualcomm.com/doc/80-70015-50/topic/qtimlvconverter.html) | <ol class="ol" id="gst-ai-segmentation__ol_j34_ddg_q1c"><br>                                    <li class="li">Receives the video stream on its sink pad.</li><br><br>                                    <li class="li">Performs the following preprocessing on the stream data.<br>                                        This preprocessing is done when the model expects<br>                                        floating-point values as input.<ol class="ol" type="a" id="gst-ai-segmentation__ol_m5z_cpr_lbc"><br>                                            <li class="li">Color conversion</li><br><br>                                            <li class="li">Scaling (up or down)</li><br><br>                                            <li class="li">Normalization</li><br><br>                                        </ol><br></li><br><br>                                    <li class="li">Converts the preprocessed video stream to a tensor stream on<br>                                        its source pad. </li><br><br>                                </ol><br><br>                                <br>The tensor stream is used for inferencing in the later stages of<br>                                    the pipeline. |
| [qtimlsnpe](https://docs.qualcomm.com/doc/80-70015-50/topic/qtimlsnpe.html) and [qtimltflite](https://docs.qualcomm.com/doc/80-70015-50/topic/qtimltflite.html) | **qtimlsnpe**: Provides ML inferencing capabilities<ul class="ul" id="gst-ai-segmentation__ul_d4s_jhs_lbc"><br>                                    <li class="li">Operates within the Qualcomm Neural Processing SDK<br>                                        runtime.</li><br><br>                                    <li class="li">Uses <span class="ph filepath">deeplabv3_resnet50.dlc</span> for image<br>                                        segmentation. </li><br><br>                                </ul><br>**qtimltflite**: Acts as an alternative option to<br>                                    qtimlsnpe.<ul class="ul" id="gst-ai-segmentation__ul_qls_lzr_lbc"><br>                                    <li class="li">Runs on the TFLite runtime.</li><br><br>                                    <li class="li">Uses<br>                                            <span class="ph filepath">deeplabv3_plus_mobilenet_quantized.tflite</span><br>                                        model for image segmentation.</li><br><br>                                </ul><br><br>The application runs on the external delegate to run the<br>                                    model using the Qualcomm Hexagon Tensor Processor.<br><br><br>After<br>                                    the inference runtime receives the tensor stream on its sink<br>                                    pad, qtimlsnpe or qtimltflite do the following:<br><ul class="ul" id="gst-ai-segmentation__ul_xvz_d3s_lbc"><br>                                    <li class="li">Run the inference.</li><br><br>                                    <li class="li">Produce a tensor stream containing the inference results,<br>                                        which is then made available on its source pad.</li><br><br>                                </ul> |
| [qtimlvsegmentation](https://docs.qualcomm.com/doc/80-70015-50/topic/qtimlvsegmentation.html) | Converts the inference tensors that it receives on its sink pad<br>                                into video formats that the multimedia plugins use for further<br>                                processing. |
| [qtivcomposer](https://docs.qualcomm.com/doc/80-70015-50/topic/qtivcomposer.html) | <ol class="ol" id="gst-ai-segmentation__ul_mr1_d1s_lbc"><br>                                    <li class="li">Composes frames by combining content from its sink pads. </li><br><br>                                    <li class="li">Pushes the GStreamer buffers containing these composed<br>                                        frames to its source pad.</li><br><br>                                </ol> |
| [Waylandsink](https://docs.qualcomm.com/doc/80-70015-50/topic/waylandsink.html) | <ol class="ol" id="gst-ai-segmentation__ol_hgj_kqr_lbc"><br>                                    <li class="li">Forwards the video stream received on its sink pad to<br>                                        Weston.</li><br><br>                                    <li class="li">Weston renders the video stream on a local display.</li><br><br>                                </ol> |

## Known issues

Model inferencing on HTP is faster than GPU.

**Parent Topic:** [AI/ML sample applications](https://docs.qualcomm.com/doc/80-70015-50/topic/ai-ml-sample-applications.html)

Last Published: Oct 27, 2025

[Previous Topic
Pose detection](https://docs.qualcomm.com/bundle/publicresource/80-70015-50/topics/gst-ai-pose-detection.md) [Next Topic
Parallel AI fusion](https://docs.qualcomm.com/bundle/publicresource/80-70015-50/topics/gst-ai-parallel-inference.md)