Hi team,
I am trying to build the efficientdet-imx project and run it on imx8mp board.
I am following this documentation:
https://github.com/nxp-imx/efficientdet-imx
But it seems that the build process for the dependencies have changed.
From the root directory of Tensorflow github, go to tensorflow/lite/tools/make and run download_dependencies.sh
Tensorflow have deprecated the use of Makefile for builds.
For Linux users, refer to https://docs.opencv.org/master/d7/d9f/tutorial_linux_install.html
Also, while following OpenCV build process we are facing the following issue:
/home/progna/Downloads/toolchains/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: ../../3rdparty/ippicv/ippicv_lnx/icv/lib/intel64/libippicv.a(ippcpufeatures.o): Relocations in generic ELF (EM: 62)
/home/progna/Downloads/toolchains/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: ../../3rdparty/ippicv/ippicv_lnx/icv/lib/intel64/libippicv.a(ippcpufeatures.o): Relocations in generic ELF (EM: 62)
/home/progna/Downloads/toolchains/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu/bin/../lib/gcc/aarch64-linux-gnu/8.3.0/../../../../aarch64-linux-gnu/bin/ld: ../../3rdparty/ippicv/ippicv_lnx/icv/lib/intel64/libippicv.a: error adding symbols: file in wrong format
collect2: error: ld returned 1 exit status
make[2]: *** [modules/core/CMakeFiles/opencv_core.dir/build.make:1465: lib/libopencv_core.so.4.9.0] Error 1
make[1]: *** [CMakeFiles/Makefile2:2196: modules/core/CMakeFiles/opencv_core.dir/all] Error 2
make: *** [Makefile:163: all] Error 2
has context menu
So, these are my queries:
GitHub - nxp-imx/efficientdet-imx: EfficientDet is a family of object detection models for which we provide basic enablement on i.MX8 NXP boards.
EfficientDet is a family of object detection models for which we provide basic enablement on i.MX8 NXP boards. - nxp-imx/efficientdet-imx