We think VIM3 C++ Demo is too complex. It is not friendly for users. So we provide a lite version. This document will help you use this lite version.

YOLOv7-Tiny is an object detection model. It uses bounding boxes to precisely draw each object in image.

Inference results on VIM3.

Inference speed test: USB camera about 216ms per frame. MIPI camera about 200ms per frame.

Download the YOLOv7 official code. WongKinYiu/yolov7

$ git clone https://github.com/WongKinYiu/yolov7

Refer README.md to create and train a YOLOv7 tiny model.

We provided a docker image which contains the required environment to convert the model.

Follow Docker official docs to install Docker: Install Docker Engine on Ubuntu.

Follow the command below to get Docker image:

docker pull numbqq/npu-vim3

$ git lfs install
$ git lfs clone https://github.com/khadas/aml_npu_sdk.git

$ cd aml_npu_sdk/acuity-toolkit/demo && ls
aml_npu_sdk/acuity-toolkit/demo$ ls
0_import_model.sh  1_quantize_model.sh  2_export_case_code.sh  data  dataset_npy.txt  dataset.txt  extractoutput.py  inference.sh  input.npy  model

After training the model, modify yolov7/models/yolo.py as follows.

diff --git a/models/yolo.py b/models/yolo.py
index 95a019c..a2e611d 100644
--- a/models/yolo.py
+++ b/models/yolo.py
@@ -144,7 +144,7 @@ class IDetect(nn.Module):
             x[i] = self.m[i](x[i])  # conv
             bs, _, ny, nx = x[i].shape  # x(bs,255,20,20) to x(bs,3,20,20,85)
-            x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
+            # x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
 
             if not self.training:  # inference
                 if self.grid[i].shape[2:4] != x[i].shape[2:4]:

Then, run export.py to convert the model to ONNX.

$ python export.py

Enter aml_npu_sdk/acuity-toolkit/demo and put yolov7_tiny.onnx into demo/model. Modify 0_import_model.sh, 1_quantize_model.sh and 2_export_case_code.sh as follows.

0_import_model.sh

#!/bin/bash
 
NAME=yolov7_tiny
ACUITY_PATH=../bin/
 
pegasus=${ACUITY_PATH}pegasus
if [ ! -e "$pegasus" ]; then
    pegasus=${ACUITY_PATH}pegasus.py
fi
 
#Onnx
$pegasus import onnx \
    --model  ./model/${NAME}.onnx \
    --output-model ${NAME}.json \
    --output-data ${NAME}.data 
 
#generate inpumeta  --source-file dataset.txt
$pegasus generate inputmeta \
	--model ${NAME}.json \
	--input-meta-output ${NAME}_inputmeta.yml \
	--channel-mean-value "0 0 0 0.0039215"  \
	--source-file dataset.txt

1_quantize_model.sh

#!/bin/bash
 
NAME=yolov7_tiny
ACUITY_PATH=../bin/
 
pegasus=${ACUITY_PATH}pegasus
if [ ! -e "$pegasus" ]; then
    pegasus=${ACUITY_PATH}pegasus.py
fi
 
#--quantizer asymmetric_affine --qtype  uint8
#--quantizer dynamic_fixed_point  --qtype int8(int16,note s905d3 not support int16 quantize) 
# --quantizer perchannel_symmetric_affine --qtype int8(int16, note only T3(0xBE) can support perchannel quantize)
$pegasus  quantize \
	--quantizer dynamic_fixed_point \
	--qtype int8 \
	--rebuild \
	--with-input-meta  ${NAME}_inputmeta.yml \
	--model  ${NAME}.json \
	--model-data  ${NAME}.data

2_export_case_code.sh

#!/bin/bash
 
NAME=yolov7_tiny
ACUITY_PATH=../bin/
 
pegasus=$ACUITY_PATH/pegasus
if [ ! -e "$pegasus" ]; then
    pegasus=$ACUITY_PATH/pegasus.py
fi
 
$pegasus export ovxlib\
    --model ${NAME}.json \
    --model-data ${NAME}.data \
    --model-quantize ${NAME}.quantize \
    --with-input-meta ${NAME}_inputmeta.yml \
    --dtype quantized \
    --optimize VIPNANOQI_PID0X88  \
    --viv-sdk ${ACUITY_PATH}vcmdtools \
    --pack-nbg-unify
 
rm -rf ${NAME}_nbg_unify
 
mv ../*_nbg_unify ${NAME}_nbg_unify
 
cd ${NAME}_nbg_unify
 
mv network_binary.nb ${NAME}.nb
 
cd ..
 
#save normal case demo export.data 
mkdir -p ${NAME}_normal_case_demo
mv  *.h *.c .project .cproject *.vcxproj BUILD *.linux *.export.data ${NAME}_normal_case_demo
 
# delete normal_case demo source
#rm  *.h *.c .project .cproject *.vcxproj  BUILD *.linux *.export.data
 
rm *.data *.quantize *.json *_inputmeta.yml

If you use VIM3L, optimize use VIPNANOQI_PID0X99.

After modifying, return to aml_npu_sdk and run convert-in-docker.sh.

If run succeed, converted model and library will generate in demo/yolov7_tiny_nbg_unify.

$ cd ../../
$ bash convert-in-docker.sh
$ cd acuity-toolkit/demo/yolov7_tiny_nbg_unify
$ ls
BUILD  main.c  makefile.linux  nbg_meta.json  vnn_global.h  vnn_post_process.c  vnn_post_process.h  vnn_pre_process.c  vnn_pre_process.h  vnn_yolov7tiny.c  vnn_yolov7tiny.h  yolov7_tiny.nb  yolov7tiny.vcxproj

Get the source code: khadas/vim3_npu_applications_lite

$ git clone https://github.com/khadas/vim3_npu_applications_lite

$ sudo apt update
$ sudo apt install libopencv-dev python3-opencv cmake

Put yolov7_tiny.nb into vim3_npu_applications_lite/yolov7_tiny_demo_x11_usb/nn_data. Replace yolov7_tiny_demo_x11_usb/vnn_yolov7tiny.c and yolov7_tiny_demo_x11_usb/include/vnn_yolov7tiny.h with your generating vnn_yolov7tiny.c and vnn_yolov7tiny.h.

# Compile
$ cd vim3_npu_applications_lite/yolov7_tiny_demo_x11_usb
$ bash build_vx.sh
$ cd bin_r_cv4
 
# usb
$ ./yolov7_tiny_demo_x11_usb -m ../nn_data/yolov7_tiny_88.nb -t usb -d /dev/video0
 
# mipi
$ ./yolov7_tiny_demo_x11_usb -m ../nn_data/yolov7_tiny_88.nb -t mipi -d /dev/video50