yolosegv8tensordec

This element can parse per-buffer inference tensors meta data generated by an upstream inference element

Example launch command:

Test image file, model file and labels file can be found here : https://gitlab.collabora.com/gstreamer/onnx-models

GST_DEBUG=yolosegv8tensordec
gst-launch-1.0 multifilesrc location=bus.jpg ! decodebin ! videoconvertscale
add-borders=1 qos=false ! 'video/x-raw,pixel-aspect-ratio=1/1' !
onnxinference execution-provider=cpu model-file=./yolov8s-seg.onnx input-image-format=chw
input-tensor-offset=0 input-tensor-scale=255.0 qos=false !
yolosegv8tensordec class-confidence-threshold=0.7 iou-threshold=0.8 max-detections=100 qos=false
! objectdetectionoverlay object-detection-outline-color=0xFF0000FF draw-labels=true !
segmentationoverlay ! videoconvert ! xvimagesink

The original repository of the Yolo is located at https://github.com/ultralytics/ultralytics. For easy experimentation, a object segmentation model based on Yolo architecture in Onnx format can be found at https://col.la/gstonnxmodelseg. This model already has the required tensor-ids embedded in the model It's also possible to embed tensor-ids into any model based on Yolo architecture to allow this tensor-decoder to decode tensors. This process is described in the Readme of this repository: https://col.la/gstonnxmodels"

GObject
    ╰──GInitiallyUnowned
        ╰──GstObject
            ╰──GstElement
                ╰──GstBaseTransform
                    ╰──GstYoloTensorDecoder
                        ╰──yolosegv8tensordec

video/x-raw:

video/x-raw: