Merchandise images dataset

Merchandise images dataset

TAGS
Image
Counterfeit detection
Bounding Box
E-commerce
AI model for monitoring counterfeits

MARQVISION is an AI based platform using machine learning to reduce the energy, time, and cost spent in detecting, reporting, and analyzing counterfeits, by automating such processes in e-commerce. Moreover, the startup plans on expanding their business to detecting illegal digital contents created by technologies such as deep-fake.

About

AI solutions for enterprises by monitoring fake products.

Datumo provides high quality data for smarter AI. As part of Datumo's Data Sponsorship Program, Datumo cooperated with Marqvision in building the following dataset.

Marqvision provides solutions for enterprises by monitoring fake products. Using DL-based image recognition, text analysis algorithms and bot-powered reporting system, Marqvision automatically discovers and takes down counterfeits in global online marketplaces to protect their brands and reputations.

Founded by Mark (Insup) Lee and DK (Do Kyung) Lee in 2019, Marqvision has reduced the time and cost consumed in monitoring and discovering counterfeits by 98 and 96 percent, respectively, compared to those when done manually. Marqvision currently monitors counterfeits in twenty-five global online marketplaces of ten different countries. Some of the clients are Amazon, Ebay, Alibaba, Taobao, Coupang, and Naver. In 2021, Marqvision has been selected as YC 21 by Y-Combinator and have raised $3.2M seed round. The startup plans on expanding their business to detecting illegal digital contents created by technologies such as deep-fake.


https://www.marqvision.com/

Testimonial

“Datumo provided us with very high quality data. We have worked on the same project with other data crowd-sourcing companies and yet Datumo's data inaccuracy rate was less than 10% compared to others. Datumo excels at setting and adhering to the timeline. They have set the timeline considering our necessities and provided us with data on time. Other partners repeatedly postponed deadlines, but Datumo delivered everything on time which helped us immensely on setting up a timeline for the development of our own AI model. Also, easy and quick communication via mobile messenger/email/Slack and so on was much appreciated.”

Dataset specification

5 categories of collected data will be shared as open datasets.

  • 25,000 items (csv file)

  • 139,467 set, 230,000 bbox item json files

Process of annotation

For more effective and accurate monitoring of counterfeits, Datumo took part in data crawling and designing the process of data preparation. Also, by meticulously pre- and post-processing data, we were able to build this project suitable for Cashmission, Datumo’s mobile crowd-sourcing platform, which immensely sped up the whole data preparation process.

Data Collection

  • Data-crawling of items according to their brands.
  • Pre-processing of collected data
  • Collection of six images per item via Cashmission
  • Pre-processing of collected images for bounding box labeling
  • Classification of items according to their names (BBox)
  • Classification of items according to their categories (BBox)
  • Identification of items categorized inappropriately (BBox)

Data were collected and labeled using Cash Mission, Datumo's crowd-sourcing platform.

'캐시미션(웹)'에서 전문 가이드 팀이 작성한 크라우드 유저들의 미션 이해를 돕기 위한 가이드
'캐시미션(앱)'에서 ‘구글 검색을 통해 제시어에 맞는 사진 6장을 찾는 미션’ 작업 진행 가이드 화면
'캐시미션(웹)'에서 박스치기 미션을 하는 이미지, 상품 사진에 박스를 그린 이미지

Sample Data

{
    "version": "4.5.6",
    "flags": {},
    "shapes": [
        {
            "label": "main",
            "points": [
                [
                    14.591055102545099,
                    85.75488015814184
                ],
                [
                    283.67926859402024,
                    200.16061279960465
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        }
    ],
    "imagePath": "[가방] Fendi_Brown fabric belt bag_1.jpg",
    "imageWidth": 300,
    "imageHeight": 300,
    "imageData": null
}
 

[가방] Fendi_Brown fabric belt bag_1.json

{
    "version": "4.5.6",
    "flags": {},
    "shapes": [
        {
            "label": "main",
            "points": [
                [
                    420.0145349784301,
                    306.56832003908687
                ],
                [
                    1072.9246920770324,
                    502.2164466097243
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        }
    ],
    "imagePath": "[선글라스] 루이비통_몽고메리 선글라스_3.jpg",
    "imageWidth": 1200,
    "imageHeight": 1200,
    "imageData": null
}
 

[선글라스] 루이비통_몽고메리 선글라스_3.json

{
    "version": "4.5.6",
    "flags": {},
    "shapes": [
        {
            "label": "main",
            "points": [
                [
                    38.42287473561406,
                    1.4988010832439613e-14
                ],
                [
                    186.46608567735996,
                    182.84812391363516
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        },
        {
            "label": "main",
            "points": [
                [
                    173.53852133666012,
                    0
                ],
                [
                    248.8768268384867,
                    163.4789319774926
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        }
    ],
    "imagePath": "[식음료] 켈로그_CHEEZ-IT  통 곡물_1.jpg",
    "imageWidth": 275,
    "imageHeight": 183,
    "imageData": null
}
 

[식음료] 켈로그_CHEEZ-IT 통 곡물_1.json

{
    "version": "4.5.6",
    "flags": {},
    "shapes": [
        {
            "label": "main",
            "points": [
                [
                    2.4580682475419318,
                    119.28860613071139
                ],
                [
                    495.80682475419314,
                    350.0578368999421
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        }
    ],
    "imagePath": "[신발] 아디다스_클라이마웜 2.0 M_1.jpg",
    "imageWidth": 500,
    "imageHeight": 500,
    "imageData": null
}

[신발] 아디다스_클라이마웜 2.0 M_1.json

{
    "version": "4.5.6",
    "flags": {},
    "shapes": [
        {
            "label": "main",
            "points": [
                [
                    208.13698642694106,
                    516.8389565713526
                ],
                [
                    1793.7000065962777,
                    2487.270618699904
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        }
    ],
    "imagePath": "[의류] 루이비통_래글런 크루넥 위드 리브 슬리브 포켓 디테일_1.jpg",
    "imageWidth": 2000,
    "imageHeight": 3000,
    "imageData": null
}
 

[의류] 루이비통_래글런 크루넥 위드 리브 슬리브 포켓 디테일_1.json

{
    "version": "4.5.6",
    "flags": {},
    "shapes": [
        {
            "label": "main",
            "points": [
                [
                    583.289104989746,
                    516.4453758245664
                ],
                [
                    1391.5134059731927,
                    1601.3590771446889
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        },
        {
            "label": "matched-category",
            "points": [
                [
                    672.3753139939392,
                    1414.7391787480897
                ],
                [
                    1274.1869729834923,
                    2709.6325113606827
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        },
        {
            "label": "unmatched-category",
            "points": [
                [
                    673.0725468051878,
                    2670.0239567082967
                ],
                [
                    872.0208883158447,
                    2904.7829996908717
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        },
        {
            "label": "unmatched-category",
            "points": [
                [
                    1086.8850971473541,
                    2667.039731585637
                ],
                [
                    1282.849213535351,
                    2901.798774568212
                ]
            ],
            "group_id": null,
            "shape_type": "rectangle",
            "labels": [],
            "flags": {}
        }
    ],
    "imagePath": "[의류] 루이비통_래글런 크루넥 위드 리브 슬리브 포켓 디테일_2.jpg",
    "imageWidth": 2000,
    "imageHeight": 3000,
    "imageData": null
}
 

[의류] 루이비통_래글런 크루넥 위드 리브 슬리브 포켓 디테일_2.json

Applications

Monitoring counterfeits in e-commerce.

CC BY-SA

Reusers are allowed to distribute, remix, adapt, and build upon the material in any medium or format, even commercially, so long as attribution is given to the creator. If you remix, adapt, or build upon the material, you must license the modified material under identical terms.

https://creativecommons.org/licenses/by-sa/3.0/deed.en

Merchandise images dataset

Merchandise images dataset