您好！已经在 Linux 系统下成功编译 lite.ai.toolkit 但是进 g++ 编译 yolox 的时候出现了如下错误： 如果不用 -I ，由于都是相对路径，又会找不到头文件，出现例如： fatal error: lite/ort/core/ort_core.h: No such file or directory

core\ort_types.h(272,1): error C2440: “初始化”: 无法从“ortcv::types::BoundingBoxType<int,double>”转换为“ortcv::types::BoundingBoxType<int,float>”

I got this error while building in Linux:

/usr/bin/ld: cannot find -lopencv_core
/usr/bin/ld: cannot find -lopencv_imgproc
/usr/bin/ld: cannot find -lopencv_imgcodecs
/usr/bin/ld: cannot find -lopencv_video
/usr/bin/ld: cannot find -lopencv_videoio
/usr/bin/ld: cannot find -lonnxruntime
collect2: error: ld returned 1 exit status

Please help! thanks!

Hi,

I succesully compiled on the Mac Os x.

trying face rec. algorithms and recognized that ONNX Runtime Version Detected Sim always same even changed the person.

ie : lite_glint_arcface.cpp model : std::string onnx_path = "../../../hub/onnx/cv/ms1mv3_arcface_r100.onnx";

person a - person b :

/var/folders/h6/7d637725049b0nf7_xqjkf640000gn/T/tmpl3pFGJ ; exit; LITEORT_DEBUG LogId: ../../../hub/onnx/cv/ms1mv3_arcface_r100.onnx =============== Input-Dims ============== input_node_dims: 1 input_node_dims: 3 input_node_dims: 112 input_node_dims: 112 =============== Output-Dims ============== Output: 0 Name: embedding Dim: 0 :1 Output: 0 Name: embedding Dim: 1 :512 [ WARN:0] global /Users/yanjunqiu/Desktop/third_party/library/opencv/modules/core/src/matrix_expressions.cpp (1334) assign OpenCV/MatExpr: processing of multi-channel arrays might be changed in the future: https://github.com/opencv/opencv/issues/16739 Default Version Detected Sim: 0.415043 Default Version Detected Dist: 1.08163 LITEORT_DEBUG LogId: ../../../hub/onnx/cv/ms1mv3_arcface_r100.onnx =============== Input-Dims ============== input_node_dims: 1 input_node_dims: 3 input_node_dims: 112 input_node_dims: 112 =============== Output-Dims ============== Output: 0 Name: embedding Dim: 0 :1 Output: 0 Name: embedding Dim: 1 :512 ONNXRuntime Version Detected Sim: 0.0349244

person-x personc

/var/folders/h6/7d637725049b0nf7_xqjkf640000gn/T/tmpzFKmvz ; exit; LITEORT_DEBUG LogId: ../../../hub/onnx/cv/ms1mv3_arcface_r100.onnx =============== Input-Dims ============== input_node_dims: 1 input_node_dims: 3 input_node_dims: 112 input_node_dims: 112 =============== Output-Dims ============== Output: 0 Name: embedding Dim: 0 :1 Output: 0 Name: embedding Dim: 1 :512 [ WARN:0] global /Users/yanjunqiu/Desktop/third_party/library/opencv/modules/core/src/matrix_expressions.cpp (1334) assign OpenCV/MatExpr: processing of multi-channel arrays might be changed in the future: https://github.com/opencv/opencv/issues/16739 Default Version Detected Sim: 0.0609607 Default Version Detected Dist: 1.37043 LITEORT_DEBUG LogId: ../../../hub/onnx/cv/ms1mv3_arcface_r100.onnx =============== Input-Dims ============== input_node_dims: 1 input_node_dims: 3 input_node_dims: 112 input_node_dims: 112 =============== Output-Dims ============== Output: 0 Name: embedding Dim: 0 :1 Output: 0 Name: embedding Dim: 1 :512 ONNXRuntime Version Detected Sim: 0.0349244

Hi,

in Mac Os x all is good. compiling succesfully . But raspi iot ubunutu has issue.

gives :

/usr/bin/ld: /home/pi/Projects/lite.ai.toolkit/build/lite.ai.toolkit/lib/liblite.ai.toolkit.so: undefined reference tocv::Mat::Mat()' ` also liblite examples same error.

undefined reference tocv::Mat::Mat()'`

when I look liblite.ai.toolkit.so with ldd: it linked to the opencv.

ldd /home/pi/Projects/lite.ai.toolkit/build/lite.ai.toolkit/lib/liblite.ai.toolkit.so linux-vdso.so.1 (0x0000007f9e95c000) libopencv_video.so.4.5 => /usr/local/lib/libopencv_video.so.4.5 (0x0000007f9e762000) libopencv_videoio.so.4.5 => /usr/local/lib/libopencv_videoio.so.4.5 (0x0000007f9e6e9000) libonnxruntime.so.1.11.0 => /home/pi/USBA/onnxruntime/build/Linux/RelWithDebInfo/libonnxruntime.so.1.11.0 (0x0000007f9dd30000) libopencv_calib3d.so.4.5 => /usr/local/lib/libopencv_calib3d.so.4.5 (0x0000007f9db7b000) libopencv_features2d.so.4.5 => /usr/local/lib/libopencv_features2d.so.4.5 (0x0000007f9dabc000) libopencv_flann.so.4.5 => /usr/local/lib/libopencv_flann.so.4.5 (0x0000007f9da50000) libopencv_imgcodecs.so.4.5 => /usr/local/lib/libopencv_imgcodecs.so.4.5 (0x0000007f9d7e6000) libopencv_imgproc.so.4.5 => /usr/local/lib/libopencv_imgproc.so.4.5 (0x0000007f9d3b2000) libopencv_core.so.4.5 => /usr/local/lib/libopencv_core.so.4.5 (0x0000007f9d036000) libstdc++.so.6 => /usr/lib/aarch64-linux-gnu/libstdc++.so.6 (0x0000007f9ce8d000) libm.so.6 => /lib/aarch64-linux-gnu/libm.so.6 (0x0000007f9cdd0000) libgcc_s.so.1 => /lib/aarch64-linux-gnu/libgcc_s.so.1 (0x0000007f9cdac000) libc.so.6 => /lib/aarch64-linux-gnu/libc.so.6 (0x0000007f9cc3a000) libdl.so.2 => /lib/aarch64-linux-gnu/libdl.so.2 (0x0000007f9cc26000) libpthread.so.0 => /lib/aarch64-linux-gnu/libpthread.so.0 (0x0000007f9cbf7000) libdc1394.so.22 => /usr/lib/aarch64-linux-gnu/libdc1394.so.22 (0x0000007f9cb75000) libgstreamer-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libgstreamer-1.0.so.0 (0x0000007f9ca20000) libgobject-2.0.so.0 => /usr/lib/aarch64-linux-gnu/libgobject-2.0.so.0 (0x0000007f9c9b9000) libglib-2.0.so.0 => /usr/lib/aarch64-linux-gnu/libglib-2.0.so.0 (0x0000007f9c886000) libgstapp-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libgstapp-1.0.so.0 (0x0000007f9c867000) libgstriff-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libgstriff-1.0.so.0 (0x0000007f9c849000) libgstpbutils-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libgstpbutils-1.0.so.0 (0x0000007f9c7ff000) libavcodec.so.58 => /usr/lib/aarch64-linux-gnu/libavcodec.so.58 (0x0000007f9b507000) libavformat.so.58 => /usr/lib/aarch64-linux-gnu/libavformat.so.58 (0x0000007f9b2a1000) libavutil.so.56 => /usr/lib/aarch64-linux-gnu/libavutil.so.56 (0x0000007f9b217000) libswscale.so.5 => /usr/lib/aarch64-linux-gnu/libswscale.so.5 (0x0000007f9b196000) librt.so.1 => /lib/aarch64-linux-gnu/librt.so.1 (0x0000007f9b17e000) libjpeg.so.62 => /usr/lib/aarch64-linux-gnu/libjpeg.so.62 (0x0000007f9b12e000) libpng16.so.16 => /usr/lib/aarch64-linux-gnu/libpng16.so.16 (0x0000007f9b0e9000) libtiff.so.5 => /usr/lib/aarch64-linux-gnu/libtiff.so.5 (0x0000007f9b05e000) libz.so.1 => /lib/aarch64-linux-gnu/libz.so.1 (0x0000007f9b031000) libtbb.so => /usr/local/lib/libtbb.so (0x0000007f9aff1000) /lib/ld-linux-aarch64.so.1 (0x0000007f9e92e000) libraw1394.so.11 => /usr/lib/aarch64-linux-gnu/libraw1394.so.11 (0x0000007f9afd4000) libusb-1.0.so.0 => /lib/aarch64-linux-gnu/libusb-1.0.so.0 (0x0000007f9afad000) libgmodule-2.0.so.0 => /usr/lib/aarch64-linux-gnu/libgmodule-2.0.so.0 (0x0000007f9af97000) libffi.so.6 => /usr/lib/aarch64-linux-gnu/libffi.so.6 (0x0000007f9af7f000) libpcre.so.3 => /lib/aarch64-linux-gnu/libpcre.so.3 (0x0000007f9af0c000) libgstbase-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libgstbase-1.0.so.0 (0x0000007f9ae8c000) libgstaudio-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libgstaudio-1.0.so.0 (0x0000007f9ae0d000) libgsttag-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libgsttag-1.0.so.0 (0x0000007f9adc1000) libgstvideo-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libgstvideo-1.0.so.0 (0x0000007f9ad1f000) libswresample.so.3 => /usr/lib/aarch64-linux-gnu/libswresample.so.3 (0x0000007f9acf8000) libvpx.so.5 => /usr/lib/aarch64-linux-gnu/libvpx.so.5 (0x0000007f9ab54000) libwebpmux.so.3 => /usr/lib/aarch64-linux-gnu/libwebpmux.so.3 (0x0000007f9ab3b000) libwebp.so.6 => /usr/lib/aarch64-linux-gnu/libwebp.so.6 (0x0000007f9aae1000) liblzma.so.5 => /lib/aarch64-linux-gnu/liblzma.so.5 (0x0000007f9aaac000) librsvg-2.so.2 => /usr/lib/aarch64-linux-gnu/librsvg-2.so.2 (0x0000007f9a5f2000) libcairo.so.2 => /usr/lib/aarch64-linux-gnu/libcairo.so.2 (0x0000007f9a4d3000) libzvbi.so.0 => /usr/lib/aarch64-linux-gnu/libzvbi.so.0 (0x0000007f9a439000) libsnappy.so.1 => /usr/lib/aarch64-linux-gnu/libsnappy.so.1 (0x0000007f9a420000) libaom.so.0 => /usr/lib/aarch64-linux-gnu/libaom.so.0 (0x0000007f9a0f8000) libcodec2.so.0.8.1 => /usr/lib/aarch64-linux-gnu/libcodec2.so.0.8.1 (0x0000007f9a087000) libgsm.so.1 => /usr/lib/aarch64-linux-gnu/libgsm.so.1 (0x0000007f9a06a000) libmp3lame.so.0 => /usr/lib/aarch64-linux-gnu/libmp3lame.so.0 (0x0000007f99fed000) libopenjp2.so.7 => /usr/lib/aarch64-linux-gnu/libopenjp2.so.7 (0x0000007f99f8d000) libopus.so.0 => /usr/lib/aarch64-linux-gnu/libopus.so.0 (0x0000007f99f2f000) libshine.so.3 => /usr/lib/aarch64-linux-gnu/libshine.so.3 (0x0000007f99f15000) libspeex.so.1 => /usr/lib/aarch64-linux-gnu/libspeex.so.1 (0x0000007f99ef0000) libtheoraenc.so.1 => /usr/lib/aarch64-linux-gnu/libtheoraenc.so.1 (0x0000007f99ead000) libtheoradec.so.1 => /usr/lib/aarch64-linux-gnu/libtheoradec.so.1 (0x0000007f99e84000) libtwolame.so.0 => /usr/lib/aarch64-linux-gnu/libtwolame.so.0 (0x0000007f99e54000) libvorbis.so.0 => /usr/lib/aarch64-linux-gnu/libvorbis.so.0 (0x0000007f99e1b000) libvorbisenc.so.2 => /usr/lib/aarch64-linux-gnu/libvorbisenc.so.2 (0x0000007f99d6b000) libwavpack.so.1 => /usr/lib/aarch64-linux-gnu/libwavpack.so.1 (0x0000007f99d36000) libx264.so.155 => /usr/lib/aarch64-linux-gnu/libx264.so.155 (0x0000007f99ae4000) libx265.so.165 => /usr/lib/aarch64-linux-gnu/libx265.so.165 (0x0000007f99838000) libxvidcore.so.4 => /usr/lib/aarch64-linux-gnu/libxvidcore.so.4 (0x0000007f99750000) libva.so.2 => /usr/lib/aarch64-linux-gnu/libva.so.2 (0x0000007f9971f000) libxml2.so.2 => /usr/lib/aarch64-linux-gnu/libxml2.so.2 (0x0000007f9956f000) libbz2.so.1.0 => /lib/aarch64-linux-gnu/libbz2.so.1.0 (0x0000007f9954b000) libgme.so.0 => /usr/lib/aarch64-linux-gnu/libgme.so.0 (0x0000007f994f5000) libopenmpt.so.0 => /usr/lib/aarch64-linux-gnu/libopenmpt.so.0 (0x0000007f9931c000) libchromaprint.so.1 => /usr/lib/aarch64-linux-gnu/libchromaprint.so.1 (0x0000007f992f9000) libbluray.so.2 => /usr/lib/aarch64-linux-gnu/libbluray.so.2 (0x0000007f9929e000) libgnutls.so.30 => /usr/lib/aarch64-linux-gnu/libgnutls.so.30 (0x0000007f990ca000) libssh-gcrypt.so.4 => /usr/lib/aarch64-linux-gnu/libssh-gcrypt.so.4 (0x0000007f99039000) libva-drm.so.2 => /usr/lib/aarch64-linux-gnu/libva-drm.so.2 (0x0000007f99026000) libva-x11.so.2 => /usr/lib/aarch64-linux-gnu/libva-x11.so.2 (0x0000007f99010000) libvdpau.so.1 => /usr/lib/aarch64-linux-gnu/libvdpau.so.1 (0x0000007f98ffc000) libX11.so.6 => /usr/lib/aarch64-linux-gnu/libX11.so.6 (0x0000007f98eb2000) libdrm.so.2 => /usr/lib/aarch64-linux-gnu/libdrm.so.2 (0x0000007f98e90000) libzstd.so.1 => /usr/lib/aarch64-linux-gnu/libzstd.so.1 (0x0000007f98df7000) libjbig.so.0 => /usr/lib/aarch64-linux-gnu/libjbig.so.0 (0x0000007f98dda000) libudev.so.1 => /lib/aarch64-linux-gnu/libudev.so.1 (0x0000007f98da6000) liborc-0.4.so.0 => /usr/lib/aarch64-linux-gnu/liborc-0.4.so.0 (0x0000007f98d17000) libsoxr.so.0 => /usr/lib/aarch64-linux-gnu/libsoxr.so.0 (0x0000007f98cb5000) libgdk_pixbuf-2.0.so.0 => /usr/lib/aarch64-linux-gnu/libgdk_pixbuf-2.0.so.0 (0x0000007f98c7e000) libgio-2.0.so.0 => /usr/lib/aarch64-linux-gnu/libgio-2.0.so.0 (0x0000007f98a9e000) libpangocairo-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libpangocairo-1.0.so.0 (0x0000007f98a80000) libpangoft2-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libpangoft2-1.0.so.0 (0x0000007f98a5a000) libpango-1.0.so.0 => /usr/lib/aarch64-linux-gnu/libpango-1.0.so.0 (0x0000007f98a01000) libfontconfig.so.1 => /usr/lib/aarch64-linux-gnu/libfontconfig.so.1 (0x0000007f989aa000) libcroco-0.6.so.3 => /usr/lib/aarch64-linux-gnu/libcroco-0.6.so.3 (0x0000007f9895e000) libpixman-1.so.0 => /usr/lib/aarch64-linux-gnu/libpixman-1.so.0 (0x0000007f988ee000) libfreetype.so.6 => /usr/lib/aarch64-linux-gnu/libfreetype.so.6 (0x0000007f9882f000) libxcb-shm.so.0 => /usr/lib/aarch64-linux-gnu/libxcb-shm.so.0 (0x0000007f9881b000) libxcb.so.1 => /usr/lib/aarch64-linux-gnu/libxcb.so.1 (0x0000007f987e4000) libxcb-render.so.0 => /usr/lib/aarch64-linux-gnu/libxcb-render.so.0 (0x0000007f987c5000) libXrender.so.1 => /usr/lib/aarch64-linux-gnu/libXrender.so.1 (0x0000007f987ac000) libXext.so.6 => /usr/lib/aarch64-linux-gnu/libXext.so.6 (0x0000007f9878c000) libogg.so.0 => /usr/lib/aarch64-linux-gnu/libogg.so.0 (0x0000007f98774000) libnuma.so.1 => /usr/lib/aarch64-linux-gnu/libnuma.so.1 (0x0000007f98754000) libicui18n.so.63 => /usr/lib/aarch64-linux-gnu/libicui18n.so.63 (0x0000007f9847f000) libicuuc.so.63 => /usr/lib/aarch64-linux-gnu/libicuuc.so.63 (0x0000007f982a4000) libicudata.so.63 => /usr/lib/aarch64-linux-gnu/libicudata.so.63 (0x0000007f968a6000) libmpg123.so.0 => /usr/lib/aarch64-linux-gnu/libmpg123.so.0 (0x0000007f96847000) libvorbisfile.so.3 => /usr/lib/aarch64-linux-gnu/libvorbisfile.so.3 (0x0000007f9682e000) libp11-kit.so.0 => /usr/lib/aarch64-linux-gnu/libp11-kit.so.0 (0x0000007f966e6000) libidn2.so.0 => /usr/lib/aarch64-linux-gnu/libidn2.so.0 (0x0000007f966b9000) libunistring.so.2 => /usr/lib/aarch64-linux-gnu/libunistring.so.2 (0x0000007f96533000) libtasn1.so.6 => /usr/lib/aarch64-linux-gnu/libtasn1.so.6 (0x0000007f96512000) libnettle.so.6 => /usr/lib/aarch64-linux-gnu/libnettle.so.6 (0x0000007f964cb000) libhogweed.so.4 => /usr/lib/aarch64-linux-gnu/libhogweed.so.4 (0x0000007f96484000) libgmp.so.10 => /usr/lib/aarch64-linux-gnu/libgmp.so.10 (0x0000007f963fa000) libgcrypt.so.20 => /lib/aarch64-linux-gnu/libgcrypt.so.20 (0x0000007f9632d000) libgssapi_krb5.so.2 => /usr/lib/aarch64-linux-gnu/libgssapi_krb5.so.2 (0x0000007f962d4000) libXfixes.so.3 => /usr/lib/aarch64-linux-gnu/libXfixes.so.3 (0x0000007f962be000) libgomp.so.1 => /usr/lib/aarch64-linux-gnu/libgomp.so.1 (0x0000007f96282000) libmount.so.1 => /lib/aarch64-linux-gnu/libmount.so.1 (0x0000007f96210000) libselinux.so.1 => /lib/aarch64-linux-gnu/libselinux.so.1 (0x0000007f961dc000) libresolv.so.2 => /lib/aarch64-linux-gnu/libresolv.so.2 (0x0000007f961b6000) libharfbuzz.so.0 => /usr/lib/aarch64-linux-gnu/libharfbuzz.so.0 (0x0000007f960b8000) libthai.so.0 => /usr/lib/aarch64-linux-gnu/libthai.so.0 (0x0000007f9609f000) libfribidi.so.0 => /usr/lib/aarch64-linux-gnu/libfribidi.so.0 (0x0000007f96072000) libexpat.so.1 => /lib/aarch64-linux-gnu/libexpat.so.1 (0x0000007f96033000) libuuid.so.1 => /lib/aarch64-linux-gnu/libuuid.so.1 (0x0000007f9601b000) libXau.so.6 => /usr/lib/aarch64-linux-gnu/libXau.so.6 (0x0000007f96008000) libXdmcp.so.6 => /usr/lib/aarch64-linux-gnu/libXdmcp.so.6 (0x0000007f95ff2000) libgpg-error.so.0 => /lib/aarch64-linux-gnu/libgpg-error.so.0 (0x0000007f95fc0000) libkrb5.so.3 => /usr/lib/aarch64-linux-gnu/libkrb5.so.3 (0x0000007f95ed5000) libk5crypto.so.3 => /usr/lib/aarch64-linux-gnu/libk5crypto.so.3 (0x0000007f95e93000) libcom_err.so.2 => /lib/aarch64-linux-gnu/libcom_err.so.2 (0x0000007f95e7f000) libkrb5support.so.0 => /usr/lib/aarch64-linux-gnu/libkrb5support.so.0 (0x0000007f95e62000) libkeyutils.so.1 => /lib/aarch64-linux-gnu/libkeyutils.so.1 (0x0000007f95e4d000) libblkid.so.1 => /lib/aarch64-linux-gnu/libblkid.so.1 (0x0000007f95de6000) libgraphite2.so.3 => /usr/lib/aarch64-linux-gnu/libgraphite2.so.3 (0x0000007f95db3000) libdatrie.so.1 => /usr/lib/aarch64-linux-gnu/libdatrie.so.1 (0x0000007f95d9b000) libbsd.so.0 => /usr/lib/aarch64-linux-gnu/libbsd.so.0 (0x0000007f95d75000)

i couldn solve

Best

我使用您的代码框架测试了一下yolox系列的推理速度，yolox_nano以外的模型推理速度都很正常，但是使用nano模型时，推理速度甚至低于yolox_s。所用的onnx文件均为利用官方coco数据集训练出来的pth文件转化得到。 我注意到yolox在定义nano模型时，有一段额外代码（./exps/default/nano.py中），如下图所示 这是否会有影响？

在BasicMNNHandler::initialize_handler中 会调用batch() channel() height()等几个函数 获取模型的输入维度信息 我这都失败了 调试显示dim里 没有任何数据 batch函数去返回dim[0]时 就崩溃了 我这用MNN是1.2.0 模型用的是你上传的网盘的模型 我试了nanodet和retinaface等模型 都不能获得维度信息 请问是是哪里的问题呢？

int main()
{
  //test_lite();
    std::string onnx_path = "/Users/also/Downloads/lite.ai.toolkit-main/hub/onnx/cv/FaceLandmark1000.onnx";
    std::string test_img_path = "/Users/also/Downloads/lite.ai.toolkit-main/examples/lite/resources/test_lite_face_landmarks_0.png";
    std::string save_img_path = "/Users/also/Downloads/lite.ai.toolkit-main/logs/test_lite_face_landmarks_1000.jpg";

    lite::cv::face::align::FaceLandmark1000 *face_landmarks_1000 =
            new lite::cv::face::align::FaceLandmark1000(onnx_path);

    lite::types::Landmarks landmarks;

    cv::VideoCapture cap;
    cv::Mat im;
    cap.open(1);
    if(! cap.isOpened())
    {
        std::cerr << "Cannot open the camera." << std::endl;
        return 0;
    }

    if( cap.isOpened()) {
        while (true) {
            cap >> im;
            //cout << "Image size: " << im.rows << "X" << im.cols << endl;
            cv::Mat img_bgr = im.clone();

            //cv::Mat img_bgr = cv::imread(test_img_path);
            face_landmarks_1000->detect(img_bgr, landmarks);

            lite::utils::draw_landmarks_inplace(img_bgr, landmarks);

            //cv::imwrite(save_img_path, img_bgr);

            std::cout << "Default Version Done! Detected Landmarks Num: "
                      << landmarks.points.size() << std::endl;


            cv::imshow("result", img_bgr);

            if((cv::waitKey(2)& 0xFF) == 'q')
                break;
        }
    }
  return 0;
}

Undefined symbols for architecture x86_64:
  "cv::imshow(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, cv::_InputArray const&)", referenced from:
      _main in test_lite_face_landmarks_1000.cpp.o
  "cv::waitKey(int)", referenced from:
      _main in test_lite_face_landmarks_1000.cpp.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[3]: *** [lite.ai.toolkit/bin/lite_face_landmarks_1000] Error 1
make[2]: *** [examples/lite/CMakeFiles/lite_face_landmarks_1000.dir/all] Error 2
make[1]: *** [examples/lite/CMakeFiles/lite_face_landmarks_1000.dir/rule] Error 2
make: *** [lite_face_landmarks_1000] Error 2

References for Windows10 users

windows下的使用可以参考以下这几个讨论(some references for windows users)

• 👉 #6
• 👉 #10
• 👉 #32
• 👉 #48
• 👉 #39
• 👉 #77

另外,是win32和system32，目前lite.ai.toolkit没有考虑32位的系统。还有就是在windows下编译完之后，需要手动把依赖库都拷贝到build/lite.ai.toolkit/lib和build/lite.ai.toolkit/bin，并且检查修改下模型文件的路径，比如说路径的反斜杠之类的。(Also, for win32 and system32, currently, lite.ai.toolkit does not consider 32-bit systems. Also, after compiling under Windows, you need to manually copy the dependent libraries to build/lite.ai.toolkit/lib and build/lite.ai.toolkit/bin, and check and modify the path of the model file, for example, The backslash of the path.)

Search issues about windows

我在win10上成功编译了此项目，运行也ok。 我现在是通过mnn来推理的，但因为cpu版本的mnn无法做到实时，所以想编译GPU版本的mnn试试，因为电脑已经装了vulkan，就编译了mnn的cpu+vulkan的库，在项目中替换了带GPU的mnn库之后，运行还是在cpu上（通过任务管理器的性能看负载，和图片的处理速度得出的） 我将MNN::ScheduleConfig schedule_config; 这里也设置成了 schedule_config.backupType = MNN_FORWARD_VULKAN; （原来是MNN_FORWARD_CPU），依然无效。 请问有办法让RVM在MNN环境下调用GPU吗？ 然后不知道MNN,ONNXRUNTIME更推荐使用哪一个呢？

如题，

我根据这两个帖子，成功的把模型转成了onnx，https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_convert_model_pytorch_specific_Convert_YOLACT.html and https://github.com/Ma-Dan/yolact/tree/onnx

问题是转换成onnx后，速度降低不少（30ms vs 125ms），主要问题出在onnx需要把资料从gpu搬回cpu上，我尝试过利用torchvision.nms处理nms，但是运算速度更慢(高于1000ms)，不知道哪里出了问题，在知乎上看见一篇文章提到，“后处理在onnx中会转换成一大坨胶水op，非常琐碎，在框架中实现效率低下”， 不知道是不是这个原因。

请问大佬你知道什么比较好的方法优化这一部分吗？谢谢

The most important thing in face recognition is to align face images after detected faces from the image. So it can correctly get the embedding code for the detected face and handle the matching function correctly. But, there is no code for face alignment in this repo. So, it can not use face recognition code of that repo.