convert pascal voc dataset to yolo format

convert_voc_to_yolo.md

Convert PascalVOC Annotations to YOLO

This script reads PascalVOC xml files, and converts them to YOLO txt files.

Note: This script was written and tested on Ubuntu. YMMV on other OS's.

Disclaimer: This code is a modified version of Joseph Redmon's voc_label.py

Instructions:

1. Place the convert_voc_to_yolo.py file into your data folder.
2. Edit the dirs array (line 8) to contain the folders where your images and xmls are located. Note: this script assumes all of your images are .jpg's (line 13).
3. Edit the classes array (line 9) to contain all of your classes.
4. Run the script. Upon running the script, each of the given directories will contain a 'yolo' folder that contains all of the YOLO txt files. A text file containing all of the image paths will be created in the cwd, for each given directory.

Make sure to put images and xml files in the root of train.Like this(image is in comment),here my folder name is VOCData and yolo folder is generated by script.

convert_voc_to_yolo.py:

import glob
import os
import pickle
import xml.etree.ElementTree as ET
from os import listdir, getcwd
from os.path import join

dirs = ['train', 'val']
classes = ['person', 'car']

def getImagesInDir(dir_path):
    image_list = []
    for filename in glob.glob(dir_path + '/*.jpg'):
        image_list.append(filename)

    return image_list

def convert(size, box):
    dw = 1./(size[0])
    dh = 1./(size[1])
    x = (box[0] + box[1])/2.0 - 1
    y = (box[2] + box[3])/2.0 - 1
    w = box[1] - box[0]
    h = box[3] - box[2]
    x = x*dw
    w = w*dw
    y = y*dh
    h = h*dh
    return (x,y,w,h)

def convert_annotation(dir_path, output_path, image_path):
    basename = os.path.basename(image_path)
    basename_no_ext = os.path.splitext(basename)[0]

    in_file = open(dir_path + '/' + basename_no_ext + '.xml')
    out_file = open(output_path + basename_no_ext + '.txt', 'w')
    tree = ET.parse(in_file)
    root = tree.getroot()
    size = root.find('size')
    w = int(size.find('width').text)
    h = int(size.find('height').text)

    for obj in root.iter('object'):
        difficult = obj.find('difficult').text
        cls = obj.find('name').text
        if cls not in classes or int(difficult)==1:
            continue
        cls_id = classes.index(cls)
        xmlbox = obj.find('bndbox')
        b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(xmlbox.find('ymin').text), float(xmlbox.find('ymax').text))
        bb = convert((w,h), b)
        out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')

cwd = getcwd()

for dir_path in dirs:
    full_dir_path = cwd + '/' + dir_path
    output_path = full_dir_path +'/yolo/'

    if not os.path.exists(output_path):
        os.makedirs(output_path)

    image_paths = getImagesInDir(full_dir_path)
    list_file = open(full_dir_path + '.txt', 'w')

    for image_path in image_paths:
        list_file.write(image_path + '\n')
        convert_annotation(full_dir_path, output_path, image_path)
    list_file.close()

    print("Finished processing: " + dir_path)

The author's code works. In VOC dataset, tag xml files and jpg images are stored separately. You need to place your xml tag file and jpg image in the same folder and then run the program to get the result of the conversion. Ensuring file paths as:
├──convert_voc_to_yolo.py
└──train
└── xxx0.xml
└── xxx0.jpg
└──test
└── xxx1.xml
└── xxx1.jpg