Tinyyolov2 and Tinyyolov3 models provided by Rdigerun are not working with other hardware options in OpenVINO

[jafet-chaves]

Any other hardware option for OpenVINO backend is not working with current models provided by RIdgerun.

Here is a sample pipeline to reproduce the issue:

gst-launch-1.0 filesrc location=Test_benchmark_video.mp4 num-buffers=600 ! decodebin ! videoconvert ! perf print-arm-load=true name=inputperf ! tee name=t t. ! videoscale ! queue ! net.sink_model t. ! queue ! net.sink_bypass tinyyolov2 backend=onnxrt_openvino name=net model-location=/home/jafet/work/devdirs/ridgerun/benchmark-onnxrt/TinyYoloV2_ONNXRT/graph_tinyyolov2.onnx backend::graph-optimization-level=0 backend::intra-num-threads=0 backend::hardware-id=GPU_FP32 net.src_bypass ! perf print-arm-load=true name=outputperf ! videoconvert ! fakesink sync=false
Setting pipeline to PAUSED ...
[WARN] 2020-08-21T14:28:36z src/ngraph/frontend/onnx_import/ops_bridge.cpp 190  Domain 'ai.onnx.ml' not recognized by nGraph
[WARN] 2020-08-21T14:28:36z src/ngraph/frontend/onnx_import/ops_bridge.cpp 190  Domain 'com.microsoft.mlfeaturizers' not recognized by nGraph
[WARN] 2020-08-21T14:28:36z src/ngraph/frontend/onnx_import/ops_bridge.cpp 190  Domain 'com.microsoft' not recognized by nGraph
[WARN] 2020-08-21T14:28:36z src/ngraph/frontend/onnx_import/ops_bridge.cpp 190  Domain 'com.microsoft.nchwc' not recognized by nGraph
[WARN] 2020-08-21T14:28:36z src/ngraph/frontend/onnx_import/ops_bridge.cpp 190  Domain 'ai.onnx.training' not recognized by nGraph
[WARN] 2020-08-21T14:28:36z src/ngraph/frontend/onnx_import/ops_bridge.cpp 190  Domain 'ai.onnx.preview.training' not recognized by nGraph
2020-08-21 08:28:39.530751336 [E:onnxruntime:, inference_session.cc:960 Initialize] Exception during initialization: /home/jafet/work/devdirs/ridgerun/onnxruntime/onnxruntime/core/providers/openvino/backends/basic_backend.cc:41 onnxruntime::openvino_ep::BasicBackend::BasicBackend(const onnx::ModelProto&, onnxruntime::openvino_ep::GlobalContext&, const onnxruntime::openvino_ep::SubGraphContext&) [OpenVINO-EP]  Exception while Loading Network for graph: OpenVINOExecutionProvider_OpenVINO-EP-subgraph_1_0Error has occured for: eltwise:add_8:0
Sizes equal or broadcast is possible(true) should be false
Invalid input shapes

Pipeline is PREROLLING ...
ERROR: from element /GstPipeline:pipeline0/GstTinyyolov2:net: Could not start the selected backend: (R2Inference Error: (Code:5) Exception during initialization: /home/jafet/work/devdirs/ridgerun/onnxruntime/onnxruntime/core/providers/openvino/backends/basic_backend.cc:41 onnxruntime::openvino_ep::BasicBackend::BasicBackend(const onnx::ModelProto&, onnxruntime::openvino_ep::GlobalContext&, const onnxruntime::openvino_ep::SubGraphContext&) [OpenVINO-EP]  Exception while Loading Network for graph: OpenVINOExecutionProvider_OpenVINO-EP-subgraph_1_0Error has occured for: eltwise:add_8:0
Sizes equal or broadcast is possible(true) should be false
Invalid input shapes

)
Additional debug info:
../gst-libs/gst/r2inference/gstvideoinference.c(434): gst_video_inference_start (): /GstPipeline:pipeline0/GstTinyyolov2:net
ERROR: pipeline doesn't want to preroll.
Setting pipeline to NULL ...
Freeing pipeline ...