Error: RGB/BGR input format specified but network input channels is not 3

Hello. I am new to deepstream and am having some difficulties getting started.

Environment and setup:

• Hardware Platform (Jetson / GPU)
Jetson TX2

• DeepStream Version
6.0

• JetPack Version (valid for Jetson only)
4.6

• TensorRT Version
8.0

• Issue Type( questions, new requirements, bugs)
I am having difficulty resolving the following error:

0:04:01.863759071 13398     0x3e23a660 ERROR                nvinfer gstnvinfer.cpp:632:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Error in NvDsInferContextImpl::preparePreprocess() <nvdsinfer_context_impl.cpp:964> [UID = 1]: RGB/BGR input format specified but network input channels is not 3
ERROR: Infer Context prepare preprocessing resource failed., nvinfer error:NVDSINFER_CONFIG_FAILED

• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)

Attempting to run an onnx model with mp4 file as data source. Attached are snippets of the config files used:

Main config file

[source0]
enable=1
#Type - 1=CameraV4L2 2=URI 3=MultiURI
type=2
uri=file://video.mp4
num-sources=1
gpu-id=0
cudadec-memtype=0
[pre-process]
enable=1
config-file=config_preprocess.txt
[primary-gie]
enable=1
gpu-id=0
infer-dims=512;512;3
batch-size=1
interval=2
gie-unique-id=1
nvbuf-memory-type=0
config-file=INFER_CONFIG.txt

Inference config:

[property]
gpu-id=0
onnx-file=/filepath/the_model.onnx
network-mode=0
#num-detected-classes=80
gie-unique-id=1
network-type=0
is-classifier=0
## 1=DBSCAN, 2=NMS, 3= DBSCAN+NMS Hybrid, 4 = None(No clustering)
cluster-mode=2
maintain-aspect-ratio=1

Pre process config:

[property]
enable=1
network-input-shape=1;512;512;3
processing-width=512
processing-height=512
tensor-data-type=0

Thanks in advance!

Hi,

RGB/BGR input format specified but network input channels is not 3

Based on the error, it seems that the input format of your model is not expected.
May I know the input dimension of your model first?

You can visualize it with the below link to find this information:
https://lutzroeder.github.io/netron/

Thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.