jacky-xbb/go-yolo

Golang binding for YOLO/Darknet recognition framework

Golang Wrapper for YOLO/Darknet Framework

Darknet is an open source neural network framework written in C and CUDA. It is fast, easy to install, and supports CPU and GPU computation.

go-yolo wraps darknet with cgo.

• darknet
• YOLO (You only look once)

Features

• Specify a GPU device

• Recognize objects from webcam/video or a image file. Equivalent to

# image
./darknet detect cfg/yolo.cfg yolo.weights data/dog.jpg
# webcom
./darknet detector demo cfg/coco.data cfg/yolo.cfg yolo.weights
# video
./darknet detector demo cfg/coco.data cfg/yolo.cfg yolo.weights /path/video.mp4

Prerequisites

Requires:

• OpenCV (version 2 or 3 is optional) to be installed on your system.

  For example, install opencv2 with Brew on MacOS:

  brew install opencv@2
  sudo chown -R $(whoami):admin /usr/local
  brew link --force opencv@2

• darknet is installed correctly on your system

  • Install this fork

  git clone https://github.com/ZanLabs/darknet.git
  cd darknet
  make OPENCV=1 # optionally GPU=1
  make install

The fork contains some workaround.

• Makefile is extended to build a static library (darknet.a).
• Makefile is extended with install and uninstall commands which install the library globally so that this module can easily find and link it.
• Support to save to a video file when detect a video.
• Support to the specific output path.

Installation

go get github.com/ZanLabs/go-yolo

Usage

Firstly prepare your cfg, data folders and download weight files from the darknet project.

Specify a GPU device:

import (
    "github.com/ZanLabs/go-yolo"
)

// If not set, default GPU #0
yolo.SetGPU(1)

Detect on a single image:

import (
    "github.com/ZanLabs/go-yolo"
)

// not save
yolo.ImageDetector(
        "./cfg/coco.data", 		// datacfg
        "./cfg/yolo.cfg",  		// cfgfile
        "./yolo.weights",  		// weightfile
        "./data/dog.jpg",  		// image that you want recognize
        0.24,					// thresh default: 0.24
        0.5)					// hierThresh default: 0.5

// save image
yolo.ImageDetector(
        "./cfg/coco.data", 		// datacfg
        "./cfg/yolo.cfg",  		// cfgfile
        "./yolo.weights",  		// weightfile
        "./data/dog.jpg",  		// image that you want recognize
        0.24,					// thresh default: 0.24
        0.5, 					// hierThresh default: 0.5
        "/PATHTO/A_IMAGE_NAME")    // ignore the suffix

Detect a video from a file:

import (
    "github.com/ZanLabs/go-yolo"
)

// not save
yolo.VideoDetector(
        "./cfg/coco.data", // datacfg
        "./cfg/yolo.cfg",  // cfgfile
        "./yolo.weights",  // weightfile
        "/path/video.mp4", // video that you want recognize
        0.24,              // thresh default: 0.24
        0.5)               // hierThresh default: 0.5

// save video
yolo.VideoDetector(
        "./cfg/coco.data", 		// datacfg
        "./cfg/yolo.cfg",  		// cfgfile
        "./yolo.weights",  		// weightfile
        "/path/video.mp4", 		// video that you want recognize
        0.24,              		// thresh default: 0.24
        0.5,               		// hierThresh default: 0.5
        "/PATHTO/A_VIDEO_NAME") // ignore the suffix

Detect video from camera:

import (
    "github.com/ZanLabs/go-yolo"
)

// not save
yolo.CameraDetector(
        "./cfg/coco.data", // datacfg
        "./cfg/yolo.cfg",  // cfgfile
        "./yolo.weights",  // weightfile
        0,                 // Camera device default: 0
        0.24,              // thresh default: 0.24
        0.5)               // hierThresh default: 0.5
 
// save frames into a dir you want
yolo.CameraDetector(
        "./cfg/coco.data", 	// datacfg
        "./cfg/yolo.cfg",  	// cfgfile
        "./yolo.weights",  	// weightfile
        0,                 	// Camera device default: 0
        0.24,              	// thresh default: 0.24
        0.5,               	// hierThresh default: 0.5
        "/PATHTO") 			// ignore the suffix