Skip to content
Branch: master
Find file History
Permalink
Type Name Latest commit message Commit time
..
Failed to load latest commit information.
imgs version 0.5.0 Jun 15, 2018
CMakeLists.txt v1.3.2 May 1, 2019
readme.md update tengine model zoo path & passwd) May 17, 2019
yolov2.cpp v1.3.2 May 1, 2019

readme.md

Yolov2 implementation with Tengine

GitHub license

This is yolov2 implementation with Tengine.

Download required models

Download the models from Tengine model zoo (psw: hhgc).

Store these files into ${Tengine_ROOT}/models/

  • yolo-voc.prototxt
  • yolo-voc.caffemodel

This caffemodel is converted from origin darknet model yolov2-voc.cfg, yolov2-voc.weight

Parameters

By default, the detected objects display with a confidence threshold of 0.24. You can change this threshold in code yolov2.cpp.

float thresh=0.24;

Build examples

cd  ${Tengine_ROOT}
make
make install

build as ${TENGINE_ROOT}/examples/readme.md

Run

  1. run yolov2 by default

    • model files are tengine/models/yolo-voc.prototxt and tengine/models/yolo-voc.caffemodel
    • test image is tengine/tests/images/ssd_dog.jpg
    cd  ${Tengine_ROOT}/examples/build/yolov2
    ./YOLOV2
    
  2. run yolov2 with other models and image

    cd  ${Tengine_ROOT}/examples/build/yolov2
    ./YOLOV2 -p proto_file -m model_file -i image_file
    
  3. show result

    Using default input image, the output image is saved as save.jpg.

    It shows detected class and its confidence score.

    save.jpg

You can’t perform that action at this time.