nnstreamer / nnstreamer

:twisted_rightwards_arrows: Neural Network (NN) Streamer, Stream Processing Paradigm for Neural Network Apps/Devices.
https://nnstreamer.ai
GNU Lesser General Public License v2.1
698 stars 172 forks source link

[Decoder] bounding-box error handling requires elaborated error messages #4458

Open myungjoo opened 4 months ago

myungjoo commented 4 months ago

There are a lot of error handling / exception handlings that do not give you enough info.

Many users are not familar with GST_DEBUG; thus let's give more info explictly.

Example: #4457

taos-ci commented 4 months ago

:octocat: cibot: Thank you for posting issue #4458. The person in charge will reply soon.

Pshubham1012 commented 1 month ago

this pipeline worked for me, just changed dimensions from 320 to 640:

gst-launch-1.0 \
v4l2src ! videoscale ! videoconvert ! video/x-raw,width=640,height=640,format=RGB,framerate=30/1,pixel-aspect-ratio=1/1 ! tee name=t \
t. ! queue ! tensor_converter ! other/tensors,num_tensors=1,types=uint8,format=static,dimensions=3:640:640:1 ! \
tensor_transform mode=arithmetic option=typecast:float32,add:0.0,div:255.0 ! \
queue leaky=2 max-size-buffers=2 ! \
tensor_filter framework=tensorflow2-lite model=yolov8s_float32.tflite custom=Delegate:XNNPACK,NumThreads:4 latency=1 ! \
other/tensors,num_tensors=1,types=float32,format=static,dimensions=8400:84:1 ! \
tensor_transform mode=transpose option=1:0:2:3 ! \
tensor_decoder mode=bounding_boxes option1=yolov8 option2=./coco-80.txt option3=0 option4=640:640 option5=640:640 ! \
video/x-raw,width=640,height=640,format=RGBA ! mix.sink_0 \
t. ! queue ! mix.sink_1 compositor name=mix sink_0::zorder=2 sink_1::zorder=1 ! videoconvert ! autovideosink

ref: https://github.com/nnstreamer/nnstreamer/commit/dabaadc553766beff030d13d393aec0477a1dba0