如何将 CSV 表转换为 python 中的 COCO 格式？

Question

我有一个 CSV 表，其中包含以下列：

column_names = ['image_id', 'xmin', 'ymin', 'width', 'height', 'xmax','ymax']

其中 xmin、ymin、xmax 和 ymax 表示包围一些 object 的边界框； 宽度和高度，图像尺寸； 和 image_id，文件名（.JPG 文件）。 由于想做object检测，所以需要把这张表转换成COCO格式。 令人惊讶的是，我在互联网上找不到这个询问的任何答案。

Answer 1

我之前遇到过同样的问题，然后我发现这段代码非常有帮助，您需要将列名更改为此列并更新 csv 文件

column_names =['filename','class','width', 'height','xmin','ymin','xmax','ymax']

然后试试这段代码

import numpy as np
import json
import pandas as pd

path = 'annotations.csv' # the path to the CSV file
save_json_path = 'traincoco.json'


data = pd.read_csv(path)

images = []
categories = []
annotations = []

category = {}
category["supercategory"] = 'none'
category["id"] = 0
category["name"] = 'None'
categories.append(category)

data['fileid'] = data['filename'].astype('category').cat.codes
data['categoryid']= pd.Categorical(data['class'],ordered= True).codes
data['categoryid'] = data['categoryid']+1
data['annid'] = data.index

def image(row):
    image = {}
    image["height"] = row.height
    image["width"] = row.width
    image["id"] = row.fileid
    image["file_name"] = row.filename
    return image

def category(row):
    category = {}
    category["supercategory"] = 'None'
    category["id"] = row.categoryid
    category["name"] = row[2]
    return category

def annotation(row):
    annotation = {}
    area = (row.xmax -row.xmin)*(row.ymax - row.ymin)
    annotation["segmentation"] = []
    annotation["iscrowd"] = 0
    annotation["area"] = area
    annotation["image_id"] = row.fileid

    annotation["bbox"] = [row.xmin, row.ymin, row.xmax -row.xmin,row.ymax-row.ymin ]

    annotation["category_id"] = row.categoryid
    annotation["id"] = row.annid
    return annotation

for row in data.itertuples():
    annotations.append(annotation(row))

imagedf = data.drop_duplicates(subset=['fileid']).sort_values(by='fileid')
for row in imagedf.itertuples():
    images.append(image(row))

catdf = data.drop_duplicates(subset=['categoryid']).sort_values(by='categoryid')
for row in catdf.itertuples():
    categories.append(category(row))

data_coco = {}
data_coco["images"] = images
data_coco["categories"] = categories
data_coco["annotations"] = annotations
json.dump(data_coco, open(save_json_path, "w"), indent=4)