mediapipe

mediapipe copied to clipboard

FaceMesh GPU Build

I have sucessfully build FaceMesh for CPU using the following command
BUILD for CPU

bazel build -c opt --define MEDIAPIPE_DISABLE_GPU=1 mediapipe/examples/desktop/face_mesh:face_mesh_cpu

But for BUILD for GPU, I got undefined reference to symbol 'nan@@GLIBC_2.2.5' error

GPU BUILD Command

bazel build -c opt --config=cuda --spawn_strategy=local --define no_aws_support=true --copt -DMESA_EGL_NO_X11_HEADERS mediapipe/examples/desktop/face_mesh:face_mesh_gpu

Error Details

ERROR: /home/turing/GPU_MediaPIPE/mediapipe/mediapipe/framework/tool/BUILD:109:10: Linking mediapipe/framework/tool/encode_as_c_string [for host] failed: (Exit 1): crosstool_wrapper_driver_is_not_gcc failed: error executing command external/local_config_cuda/crosstool/clang/bin/crosstool_wrapper_driver_is_not_gcc @bazel-out/host/bin/mediapipe/framework/tool/encode_as_c_string-2.params /usr/bin/ld: bazel-out/host/bin/external/com_google_absl/absl/strings/libstrings.a(charconv.o): undefined reference to symbol 'nan@@GLIBC_2.2.5' //lib/x86_64-linux-gnu/libm.so.6: error adding symbols: DSO missing from command line collect2: error: ld returned 1 exit status Target //mediapipe/examples/desktop/face_mesh:face_mesh_gpu failed to build Use --verbose_failures to see the command lines of failed build steps. INFO: Elapsed time: 100.569s, Critical Path: 59.76s INFO: 820 processes: 209 internal, 611 local. FAILED: Build did NOT complete successfully

I have add cc_binary( ... linkopts = ["-lm"], ) in BUILD file [/mediapipe/examples/desktop/face_mesh/BUILD]

cc_binary(
    name = "face_mesh_gpu",
    data = ["//mediapipe/modules/face_landmark:face_landmark_with_attention.tflite"],
    deps = [
        "//mediapipe/examples/desktop:demo_run_graph_main_gpu",
        "//mediapipe/graphs/face_mesh:desktop_live_gpu_calculators",
    ],
   linkopts = ["-lm"],
)

following the solution, SOLUTION

But still got the same error

Hi @robi56 , it's good that the GPU path works for facemesh when refine_landmarks is false, that means the mediapipe GPU configuration/setup is fine.

the refine_landmarks = true option actually runs a different ML model (here)

it seems that model has some op that isn't compatible with the tensorflow gpu delegate.

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you.

Closing as stale. Please reopen if you'd like to work on this further.

Are you satisfied with the resolution of your issue? Yes No