# Daisy chain detection and pose detection using Python

Source: [https://docs.qualcomm.com/doc/80-70022-50/topic/daisy-chain-detection-and-pose-detection-using-python.html](https://docs.qualcomm.com/doc/80-70022-50/topic/daisy-chain-detection-and-pose-detection-using-python.html)

The **gst-ai-daisychain-detection-pose.py** application allows you to perform
        cascaded object detection and pose detection with input from a camera, file source, or an
        RTSP stream.

Figure : gst-ai-daisychain-detection-pose.py pipeline
            
            <?xml version="1.0" encoding="UTF-8"?>
<svg id="Layer_1" data-name="Layer 1" xmlns="http://www.w3.org/2000/svg" width="1270.322357177734375" height="543.212188720703125" viewbox="0 0 1270.322357177734375 543.212188720703125">
  <g>
    <rect x=".499908447265625" y=".500213623046875" width="1269.322265625" height="542.2119140625" rx="7.5" ry="7.5" style="fill: #fafafa;"></rect>
    <path d="M1262.322357177734375,1c3.85980224609375,0,7,3.140167236328125,7,7v527.212188720703125c0,3.85980224609375-3.14019775390625,7-7,7H8c-3.859832763672784,0-7-3.14019775390625-7-7V8c0-3.859832763671875,3.140167236327216-7,7-7h1254.322357177734375M1262.322357177734375,0H8C3.5816650390625,0,0,3.581832885742188,0,8v527.212188720703125c0,4.4183349609375,3.5816650390625,8,8,8h1254.322357177734375c4.41815185546875,0,8-3.5816650390625,8-8V8c0-4.418167114257812-3.58184814453125-8-8-8h0Z" style="fill: #d2d7e1;"></path>
  </g>
  <g>
    <rect x="147.641433317429801" y="20.834238745659604" width="536.179735718712436" height="205.848021655063349" rx="7.999999999999999" ry="7.999999999999999" style="fill: none; stroke: #505050; stroke-miterlimit: 10; stroke-width: 2px;"></rect>
    <text transform="translate(402.448545455932617 47.506492614746094)" style="font-size: 16px;"><tspan x="0" y="0">Inferbin</tspan></text>
    <rect x="169.034748641877741" y="65.288945076274103" width="60.000000000001819" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(187.94500732421875 99.964691162109375)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">tee</tspan></text>
    <rect x="752.603561992684263" y="407.212270043432" width="60" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(771.512969970703125 441.8885498046875)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">tee</tspan></text>
    <rect x="250.4280639663275" y="65.288945076274103" width="130.606474420916129" height="60" rx="3.999999999999999" ry="3.999999999999999" style="fill: #2a2aea;"></rect>
    <text transform="translate(260.258575439453125 99.964515686035156)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">qtimlvconverter</tspan></text>
    <rect x="34.580211902747578" y="65.288945076274103" width="100" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(59.7286376953125 99.618644714355469)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">Source</tspan></text>
    <rect x="722.603561992684263" y="318.334848694285029" width="120" height="60" rx="4" ry="4" style="fill: #2a2aea;"></rect>
    <text transform="translate(754.517242431640625 352.66455078125)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">qtivsplit</tspan></text>
    <rect x="402.427853711692478" y="65.288945076274103" width="120" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(429.037264347076416 99.964515686035156)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">Inference</tspan></text>
    <rect x="543.821169036141328" y="65.288945076274103" width="120.000000000000909" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(559.039688110351562 99.618644714355469)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">Postprocess</tspan></text>
    <rect x="543.821169036144056" y="146.682260400722953" width="119.999999999998181" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(563.274215698242188 181.357888221740723)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">metamuxer</tspan></text>
    <g>
      <line x1="134.641433317429801" y1="95.288945076274103" x2="167.641433317429801" y2="95.288945076274103" style="fill: none; stroke: #fff; stroke-miterlimit: 10; stroke-width: 3px;"></line>
      <g>
        <line x1="134.641433715820312" y1="95.2889404296875" x2="161.90045166015625" y2="95.2889404296875" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
        <polygon points="160.733261108398438 99.27801513671875 167.641433715820312 95.2889404296875 160.733261108398438 91.299896240235284 160.733261108398438 99.27801513671875"></polygon>
      </g>
    </g>
    <g>
      <line x1="229.034751892089844" y1="95.2889404296875" x2="244.687080383300781" y2="95.2889404296875" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="243.519889831542969 99.27801513671875 250.428062438964844 95.2889404296875 243.519889831542969 91.299896240235284 243.519889831542969 99.27801513671875"></polygon>
    </g>
    <g>
      <line x1="381.034538269042969" y1="95.2889404296875" x2="396.686870574951172" y2="95.2889404296875" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="395.519676208496094 99.27801513671875 402.4278564453125 95.2889404296875 395.519676208496094 91.299896240235284 395.519676208496094 99.27801513671875"></polygon>
    </g>
    <g>
      <line x1="522.4278564453125" y1="95.2889404296875" x2="538.080184936523438" y2="95.2889404296875" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="536.912994384765625 99.27801513671875 543.8211669921875 95.2889404296875 536.912994384765625 91.299896240235284 536.912994384765625 99.27801513671875"></polygon>
    </g>
    <g>
      <line x1="603.8211669921875" y1="125.2889404296875" x2="603.8211669921875" y2="140.9412841796875" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="599.83209228515625 139.774078369140625 603.8211669921875 146.6822509765625 607.81024169921875 139.774078369140625 599.83209228515625 139.774078369140625"></polygon>
    </g>
    <g>
      <polyline points="199.034744262695312 125.2889404296875 199.034744262695312 176.6822509765625 538.080184936523438 176.6822509765625" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></polyline>
      <polygon points="536.912994384765625 180.67132568359375 543.8211669921875 176.6822509765625 536.912994384765625 172.693206787110284 536.912994384765625 180.67132568359375"></polygon>
    </g>
    <rect x="147.641433317429801" y="281.364248388366832" width="536.179735718712436" height="205.848021655065168" rx="7.999999999999999" ry="7.999999999999999" style="fill: none; stroke: #505050; stroke-miterlimit: 10; stroke-width: 2px;"></rect>
    <text transform="translate(402.448545455932617 308.0364990234375)" style="font-size: 16px;"><tspan x="0" y="0">Inferbin</tspan></text>
    <rect x="169.034748641877741" y="325.818954718981331" width="60.000000000001819" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(187.94500732421875 360.49468994140625)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">tee</tspan></text>
    <rect x="250.4280639663275" y="325.818954718981331" width="130.606474420916129" height="60" rx="3.999999999999999" ry="3.999999999999999" style="fill: #2a2aea;"></rect>
    <text transform="translate(260.258575439453125 360.494522094726562)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">qtimlvconverter</tspan></text>
    <rect x="866.720171619082976" y="318.334848694285029" width="120" height="60" rx="4" ry="4" style="fill: #2a2aea;"></rect>
    <text transform="translate(877.860870361328125 353.010421752929688)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">qtivcomposer</tspan></text>
    <rect x="1015.742082481978287" y="318.334848694285029" width="120" height="60" rx="4" ry="4" style="fill: #2a2aea;"></rect>
    <text transform="translate(1041.179718017578125 353.010421752929688)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">qtioverlay</tspan></text>
    <rect x="402.427853711692478" y="325.818954718981331" width="120" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(429.037264347076416 360.494522094726562)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">Inference</tspan></text>
    <text transform="translate(363.209926605224609 256.869094848632812)" style="font-size: 16px;"><tspan x="0" y="0">video/x-rawz</tspan></text>
    <text transform="translate(692.540283203125 431.39202880859375)" style="font-size: 16px;"><tspan x="0" y="0">video/</tspan><tspan x="3.18359375" y="20">x-raw</tspan></text>
    <rect x="543.821169036141328" y="325.818954718981331" width="120.000000000000909" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(559.039688110351562 360.148651123046875)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">Postprocess</tspan></text>
    <rect x="543.821169036144056" y="407.212270043432" width="119.999999999998181" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
    <text transform="translate(563.274215698242188 441.887908935546875)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">metamuxer</tspan></text>
    <g>
      <line x1="229.034751892089844" y1="355.818939208984375" x2="244.687080383300781" y2="355.818939208984375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="243.519889831542969 359.808013916015625 250.428062438964844 355.818939208984375 243.519889831542969 351.829864501953125 243.519889831542969 359.808013916015625"></polygon>
    </g>
    <g>
      <line x1="381.034538269042969" y1="355.818939208984375" x2="396.686870574951172" y2="355.818939208984375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="395.519676208496094 359.808013916015625 402.4278564453125 355.818939208984375 395.519676208496094 351.829864501953125 395.519676208496094 359.808013916015625"></polygon>
    </g>
    <g>
      <line x1="522.4278564453125" y1="355.818939208984375" x2="538.080184936523438" y2="355.818939208984375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="536.912994384765625 359.808013916015625 543.8211669921875 355.818939208984375 536.912994384765625 351.829864501953125 536.912994384765625 359.808013916015625"></polygon>
    </g>
    <g>
      <line x1="842.60357666015625" y1="331.720123291015625" x2="860.979217529296875" y2="331.720123291015625" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="859.811981201171875 335.70916748046875 866.720184326171875 331.720123291015625 859.811981201171875 327.731048583984375 859.811981201171875 335.70916748046875"></polygon>
    </g>
    <g>
      <line x1="842.60357666015625" y1="349.222747802734375" x2="860.979217529296875" y2="349.222747802734375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="859.811981201171875 353.211761474609375 866.720184326171875 349.222747802734375 859.811981201171875 345.233673095703125 859.811981201171875 353.211761474609375"></polygon>
    </g>
    <g>
      <line x1="986.720184326171875" y1="349.222747802734375" x2="1008.696441650390625" y2="349.222747802734375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="1007.529205322265625 353.211761474609375 1014.437408447265625 349.222747802734375 1007.529205322265625 345.233673095703125 1007.529205322265625 353.211761474609375"></polygon>
    </g>
    <line x1="1075.742082481978287" y1="378.334848694285029" x2="1075.742082481978287" y2="389.027399006090491" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
    <g>
      <line x1="842.60357666015625" y1="366.725311279296875" x2="860.979217529296875" y2="366.725311279296875" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="859.811981201171875 370.714385986328125 866.720184326171875 366.725311279296875 859.811981201171875 362.736236572265625 859.811981201171875 370.714385986328125"></polygon>
    </g>
    <g>
      <line x1="603.8211669921875" y1="385.818939208984375" x2="603.8211669921875" y2="401.471282958984375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="599.83209228515625 400.304107666015625 603.8211669921875 407.212249755859375 607.81024169921875 400.304107666015625 599.83209228515625 400.304107666015625"></polygon>
    </g>
    <g>
      <line x1="782.603546142578125" y1="407.212249755859375" x2="782.60357666015625" y2="384.075836181640625" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
      <polygon points="786.592620849609375 385.243011474609375 782.60357666015625 378.334869384765625 778.614501953125 385.243011474609375 786.592620849609375 385.243011474609375"></polygon>
    </g>
    <g>
      <polyline points="199.034744262695312 385.818939208984375 199.034744262695312 437.212249755859375 538.080184936523438 437.212249755859375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></polyline>
      <polygon points="536.912994384765625 441.201324462890625 543.8211669921875 437.212249755859375 536.912994384765625 433.223236083984375 536.912994384765625 441.201324462890625"></polygon>
    </g>
    <g>
      <polyline points="600.220178356661563 206.682260400722953 600.220178356661563 264.099451355417841 199.03474864187956 264.099451355417841 199.03474864187956 325.818954718981331" style="fill: none; stroke: #fff; stroke-miterlimit: 10; stroke-width: 3px;"></polyline>
      <g>
        <polyline points="600.220184326171875 206.6822509765625 600.220184326171875 264.099456787109375 199.034744262695312 264.099456787109375 199.034744262695312 320.077972412109375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></polyline>
        <polygon points="195.045684814452216 318.9107666015625 199.034744262695312 325.8189697265625 203.023818969726562 318.9107666015625 195.045684814452216 318.9107666015625"></polygon>
      </g>
    </g>
    <g>
      <line x1="663.821169036142237" y1="437.212270043432" x2="752.603561992684263" y2="437.212270043432" style="fill: none; stroke: #fff; stroke-miterlimit: 10; stroke-width: 3px;"></line>
      <g>
        <line x1="663.8211669921875" y1="437.212249755859375" x2="746.862579345703125" y2="437.212249755859375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
        <polygon points="745.69537353515625 441.201324462890625 752.60357666015625 437.212249755859375 745.69537353515625 433.223236083984375 745.69537353515625 441.201324462890625"></polygon>
      </g>
    </g>
    <g>
      <polyline points="812.60357666015625 437.212249755859375 926.720184326171875 437.212249755859375 926.720184326171875 384.075836181640625" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></polyline>
      <polygon points="930.709259033203125 385.243011474609375 926.720184326171875 378.334869384765625 922.731109619140625 385.243011474609375 930.709259033203125 385.243011474609375"></polygon>
    </g>
    <g>
      <g>
        <text transform="translate(1062.1356201171875 519.546844482421875)" style="font-family: Roboto-Regular, Roboto; font-size: 14px;"><tspan x="0" y="0">Qualcomm </tspan></text>
        <rect x="1041.884597813968867" y="507.455352783203125" width="16" height="16" rx="2" ry="2" style="fill: #2a2aea;"></rect>
      </g>
      <g>
        <text transform="translate(1160.7174072265625 519.546844482421875)" style="font-family: Roboto-Regular, Roboto; font-size: 14px;"><tspan x="0" y="0">Open source</tspan></text>
        <rect x="1140.466379226822028" y="507.455352783203125" width="16" height="16" rx="2" ry="2" style="fill: #007884;"></rect>
      </g>
    </g>
    <g>
      <g>
        <rect x="1125.742082481978287" y="407.212270043432" width="110" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
        <text transform="translate(1135.753936767578125 441.8880615234375)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">Waylandsink</tspan></text>
      </g>
      <g>
        <rect x="1035.742082481978287" y="407.212270043432" width="80" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
        <text transform="translate(1050.308563232421875 441.8880615234375)" style="fill: #fff; font-size: 16px;"><tspan x="0" y="0">filesink</tspan></text>
      </g>
      <g>
        <rect x="945.742082481978287" y="407.212270043432" width="80" height="60" rx="4" ry="4" style="fill: #007884;"></rect>
        <text transform="translate(956.098114013671875 441.8880615234375)" style="fill: #fff; font-family: Roboto-Regular, Roboto; font-size: 14px;"><tspan x="0" y="0">qtirtspbin</tspan></text>
      </g>
      <g>
        <polyline points="985.742095947265625 401.032562255859375 985.742095947265625 389.027374267578125 1180.742095947265625 389.027374267578125 1180.742095947265625 400.471282958984375" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></polyline>
        <polygon points="989.731170654296875 399.865386962890625 985.742095947265625 406.773529052734375 981.753021240234375 399.865386962890625 989.731170654296875 399.865386962890625"></polygon>
        <polygon points="1176.753021240234375 399.304107666015625 1180.742095947265625 406.212249755859375 1184.731170654296875 399.304107666015625 1176.753021240234375 399.304107666015625"></polygon>
      </g>
      <g>
        <line x1="1075.742095947265625" y1="401.471282958984375" x2="1075.742095947265625" y2="389.027374267578125" style="fill: none; stroke: #000; stroke-miterlimit: 10;"></line>
        <polygon points="1079.731170654296875 400.304107666015625 1075.742095947265625 407.212249755859375 1071.753021240234375 400.304107666015625 1079.731170654296875 400.304107666015625"></polygon>
      </g>
    </g>
  </g>
</svg>

For information about the plugins used in this pipeline, see [Pipeline flow](https://docs.qualcomm.com/doc/80-70022-50/topic/daisy-chain-detection-and-pose-detection-using-python.html#daisy-chain-detection-and-pose-detection-using-python__section_pqq_1ny_kbc).

## Run the application on the target device

1. Ensure that you complete the [Prerequisites](https://docs.qualcomm.com/doc/80-70022-50/topic/prerequisites-for-python-sample-applications.html).
2. Rename the label files on the
                        EVK:

        cp /etc/labels/yolov8.json /etc/labels/yolox.jsonCopy to clipboard

Note: For Ubuntu Server, run the above command with
                            `sudo`.
3. Run the use cases:
    - Input from a
                            camera:

            gst-daisychain-detection-pose.py  --cameraCopy to clipboard
    - Input from a
                            file:

            gst-daisychain-detection-pose.py  --file /etc/media/video.mp4Copy to clipboard
    - Input from an RTSP
                            stream:

            gst-daisychain-detection-pose.py  --rtsp "rtsp://<ip>:<port>/<stream>"Copy to clipboard

Note: If a drop in performance is observed, you can
                        use YOLOv8 LiteRT model. For YOLOv8 export instructions, see Step 6 in [Prerequisites](https://docs.qualcomm.com/doc/80-70022-50/topic/download-model-and-label-files.html).
4. To display the available help options, run the following
                    command:

        gst-daisychain-detection-pose.py -hCopy to clipboard

## Expected output

Figure : Expected output for gst-ai-daisychain-detection-pose.py application
                
                ![](data:image/png;base64,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)

## Pipeline flow

The following table lists the plugins used in the daisy chain detection and
                    pose estimation pipeline:| Plugin | Description |
| --- | --- |
| Camera source:[qtiqmmfsrc](https://docs.qualcomm.com/doc/80-70022-50/topic/qtiqmmfsrc.html) | <ul class="ul" id="daisy-chain-detection-and-pose-detection-using-python__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="daisy-chain-detection-and-pose-detection-using-python__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="daisy-chain-detection-and-pose-detection-using-python__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> |
| [qtimetamux](https://docs.qualcomm.com/doc/80-70022-50/topic/qtimetamux.html) | Multiplexes the stream. |
| h264parse | Parses the H.264 video. |
| [v4l2h264dec](https://docs.qualcomm.com/doc/80-70022-50/topic/v4l2h264dec.html) | Decodes the H.264 video stream. |
| [qtivsplit](https://docs.qualcomm.com/doc/80-70022-50/topic/qtivsplit.html) | Crops full frame into smaller frames based on the detected<br>                                bounding boxes detected (maximum 4). |
| [qtimlvconverter](https://docs.qualcomm.com/doc/80-70022-50/topic/qtimlvconverter.html) | Used by AI processing stream for preprocessing:<ol class="ol" id="daisy-chain-detection-and-pose-detection-using-python__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="daisy-chain-detection-and-pose-detection-using-python__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>The tensor stream is used for inferencing in the later<br>                                    stages of the pipeline. |
| [qtimltflite](https://docs.qualcomm.com/doc/80-70022-50/topic/qtimltflite.html) | Runs on LiteRT and uses the yolov5.tflite model for object<br>                                    detection and inception\_v3\_quantized.tflite<br>                                    for classification.<br><ol class="ol" id="daisy-chain-detection-and-pose-detection-using-python__ol_l2x_zjq_nbc"><br>                                    <li class="li">After the inference runtime receives the tensor stream on<br>                                        its sink pad, it runs the inference.</li><br><br>                                    <li class="li">Produces a tensor stream with the inference results on its<br>                                        source pad.</li><br><br>                                </ol> |
| qtimlpostprocess | Handles inference results from any detection model.<ol class="ol" id="daisy-chain-detection-and-pose-detection-using-python__ol_ol3_dky_kbc"><br>                                    <li class="li">Applies a threshold to the chosen number of results.</li><br><br>                                    <li class="li">Loads the YOLOv8 module. </li><br><br>                                    <li class="li">Generates results in the form of video frames with<br>                                        classification labels.</li><br><br>                                    <li class="li">Produces video frames with only bounding boxes that can be<br>                                        cropped. </li><br><br>                                </ol> |
| qtimlpostprocess | <ul class="ul" id="daisy-chain-detection-and-pose-detection-using-python__ul_jk5_b1s_lbc"><br>                                    <li class="li">Applies a threshold to the chosen number of results. </li><br><br>                                    <li class="li">Loads corresponding modules for various pose detection<br>                                        models. <p class="p">In the use case described in this section,<br>                                            qtimlpostprocess does the following:</p><ol class="ol" id="daisy-chain-detection-and-pose-detection-using-python__ol_w5v_2qr_lbc"><br>                                            <li class="li">Loads the HRNet module.</li><br><br>                                            <li class="li">Produces results in the form of video frames with<br>                                                drawn poses. </li><br><br>                                            <li class="li">Sends the results to the sink pad of qtivcomposer<br>                                                for further processing or display.</li><br><br>                                        </ol><br></li><br><br>                                </ul> |
| [qtivcomposer](https://docs.qualcomm.com/doc/80-70022-50/topic/qtivcomposer.html) | <ol class="ol" id="daisy-chain-detection-and-pose-detection-using-python__ol_y1d_hqr_lbc"><br>                                    <li class="li">Composes frames by combining the contents from its sink<br>                                        pads.</li><br><br>                                    <li class="li">Pushes the GStreamer buffers containing the composed frames<br>                                        onto its source pad.</li><br><br>                                </ol> |
| [Waylandsink](https://docs.qualcomm.com/doc/80-70022-50/topic/waylandsink.html) | <ol class="ol" id="daisy-chain-detection-and-pose-detection-using-python__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> |

## Sample model and label files

Table : Sample model and label files for gst-ai-daisychain-detection-pose

| Runtime | Model files | Label files |
| :--- | :--- | :--- |
| LiteRT | <ul class="ul" id="daisy-chain-detection-and-pose-detection-using-python__d22e84"><br>                                        <li class="li">Detection:<br>                                            <var class="keyword varname">yolox_quantized.tflite</var></li><br><br>                                        <li class="li">Pose:<br>                                            <var class="keyword varname">hrnet_pose_quantized.tflite</var></li><br><br>                                    </ul> | <ul class="ul" id="daisy-chain-detection-and-pose-detection-using-python__d22e100"><br>                                        <li class="li">Detection: <var class="keyword varname">yolox.json</var></li><br><br>                                        <li class="li">Pose:<ul class="ul" id="daisy-chain-detection-and-pose-detection-using-python__d22e109"><br>                                                <li class="li"><var class="keyword varname">hrnet_pose.json</var></li><br><br>                                                <li class="li"><var class="keyword varname">hrnet_settings.json</var></li><br><br>                                            </ul><br></li><br><br>                                    </ul> |

## Related information

[Daisy chain detection and pose estimation](https://docs.qualcomm.com/doc/80-70022-50/topic/daisy-chain-detection-and-pose-detection.html)

**Parent Topic:** [Run Python-based applications](https://docs.qualcomm.com/doc/80-70022-50/topic/python-sample-applications.html)

Last Published: Feb 20, 2026

[Previous Topic
Parallel inference using Python](https://docs.qualcomm.com/bundle/publicresource/80-70022-50/topics/parallel-inference-using-python.md) [Next Topic
Customize a Python sample application](https://docs.qualcomm.com/bundle/publicresource/80-70022-50/topics/customize-a-python-sample-application.md)