I am trying to run RVM Matting with deepstream sample application. It has more than one input layer

My current deepstream pipeline throws following error:

0 INPUT kFLOAT src 3x720x1280
1 INPUT kFLOAT r1i 1x1x1
2 INPUT kFLOAT r2i 1x1x1
3 INPUT kFLOAT r3i 1x1x1
4 INPUT kFLOAT r4i 1x1x1
5 OUTPUT kFLOAT r4o 64x18x32
6 OUTPUT kFLOAT r3o 40x36x64
7 OUTPUT kFLOAT r2o 20x72x128
8 OUTPUT kFLOAT r1o 16x144x256
9 OUTPUT kFLOAT fgr 3x720x1280
10 OUTPUT kFLOAT pha 1x720x1280

0:01:56.926524755 35 0x55ffaaa2ba70 WARN nvinfer gstnvinfer.cpp:677:gst_nvinfer_logger: NvDsInferContext[UID 1]: Warning from NvDsInferContextImpl::initNonImageInputLayers() <nvdsinfer_context_impl.cpp:1487> [UID = 1]: More than one input layers but custom initialization function not implemented
0:01:56.926542595 35 0x55ffaaa2ba70 ERROR nvinfer gstnvinfer.cpp:674:gst_nvinfer_logger: NvDsInferContext[UID 1]: Error in NvDsInferContextImpl::initialize() <nvdsinfer_context_impl.cpp:1300> [UID = 1]: Failed to initialize non-image input layers
0:01:56.944899654 35 0x55ffaaa2ba70 WARN nvinfer gstnvinfer.cpp:888:gst_nvinfer_start: error: Failed to create NvDsInferContext instance

Can you help with how to handle more than one input in deep stream nvinfer module?

Please provide information of your pipeline and complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• 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)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
• The pipeline being used

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks

Current default gst-nvinfer implementation only support one input layer model or some special multiple input layers model with only one image input layer and the extra input layers are just some fixed values(please refer to /opt/nvidia/deepstream/deepstream/sources/objectDetector_FasterRCNN ).

If the 4 extra layers are not fixed value input layers, the gst-nvinfer plugin and the nvinfer library are open source, you can modify the code to support your multiple input layers.

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