Collectibles Recognition
In order to get access to the Collectibles Recognition service, please register at https://app.ximilar.com. In the app pick "Ready-To-Use Image Recognition" on the main dashboard and then Collectibles Recognition. This service is available only in Business and Professional pricing plans. A public demo is located at demo.ximilar.com.
This page describes API of Collectibles Recognition service for images. With this service you can analyse photos of collectible items via AI. The API method first detect all collectible items on a photo and then recognizes their types (Stamp, Coin, Banknote, Comics, Card, Slab Label). For detected Cards (Trading Card Game or Sport Card) and Comics (Manga included), the service can identify name, set, series, card number, year of release and other fields. For some cards/games it also return IDs for TCG Player. It can also read information from the Slab Labels or get the prices of the collector items.
If you are looking for a more detailed analysis of collector items, e.g.:
- you are missing a specific feature, items or game and would like to add it to this endpoint,
- or you want visual search or identification of the stamp, banknotes, coins, comics, posters, vinyl covers, cards, etc.,
- or you want visual search on your own database of collectors items
Then please contact us at care@ximilar.com and we can create a customized soluton for you.
This service also supports asynchronous API with webhook. If you need to process large number of requests per day or month (millions of requests) and don't need to get the results immediately then just use our asynchronous requests with webhooks.
Endpoints
This service API has several api endpoints running at URLs, each of them are suitable for different use case and also are priced different:
https://api.ximilar.com/collectibles/v2/process (for localization and tags)
https://api.ximilar.com/collectibles/v2/analyze (for complete analysis)
https://api.ximilar.com/collectibles/v2/tcg_id (for trading card games)
https://api.ximilar.com/collectibles/v2/sport_id (for sport cards)
https://api.ximilar.com/collectibles/v2/comics_id (for comics magazines & books & manga)
https://api.ximilar.com/collectibles/v2/card_ocr_id (for OCR on cards)
https://api.ximilar.com/collectibles/v2/slab_id (for graded slabs - complex analysis)
https://api.ximilar.com/collectibles/v2/detect (for localization)
https://api.ximilar.com/collectibles/v2/slab_grade (for graded slabs - simple analysis)
TCG Identification
Given a list of image records, this method returns identification for largest Trading Card Game card that was detected on image. This service is able to further analyze and identify the trading card.
This service can identify cards from more than 15 games including Pokemon, Yugioh!, Magic The Gathering, OnePiece, Lorcana. The Full list is available on Taxonomy page. It is also able to recognize Alphabet/Language, and visual features like Holo/Foil, etc. If the input card is from one of mentioned games, then we are able to able to identify the card set, set_code, series, card_number, full_name, year of release and other fields. For some of these cards and games we have links to tcgplayer.com, scryfall.com and other sites. It can also detects and analyze the graded Slab Label (getting certificate number, grade value..).
If you need to distinguish international/latin alphabet language (English, German, Portuguese, Spanish, ...) send lang: true and in returned _ocr field you will find a proper language identification. For Pokemon and One Piece, we are able to properly identify english, chinese (simplified) and japanese variants of the cards.
Required attributes
- Name
records- Type
- dict
- Description
A batch of json records (max 10), one record is representation of an image and it's defined by
_urlor_base64.
Optional attributes
- Name
lang- Type
- boolean
- Default
- Default:False
- Description
Set true if you need to distinguish language of trading card in latin alphabet, otherwise take the language information from _tags.
- Name
slab_id- Type
- boolean
- Default
- Default:False
- Description
If set to true, then the slab identification will be done via OCR and AI, default False. This will analyse only one slab label on the image and also count additional API credits according to /v2/slab_id.
- Name
slab_grade- Type
- boolean
- Default
- Default:False
- Description
If set to true, then AI system will analyse the company and grade on the image, default False. This will also count additional API credits according to /v2/slab_grade.
- Name
analyze_all- Type
- boolean
- Default
- Default:False
- Description
If specified it will analyze all cards on image (only 1 record/image is allowed), the perfect use case is a card binder. The number of credits is per number of analysed cards on image instead of records (cards count * credit price of endpoint).
- Name
pricing- Type
- boolean
- Default
- Default:False
- Description
If set to true, then system will try to return the prices, default False. This will also count additional API credits according to pricing operation. The API credits will not be counted if the system return no prices.
Returns
HTTP error code 2XX, if the method was OK and other HTTP error code, if the method failed. Body of the response is a JSON object (map) with the following fields:
- Name
records- Type
- dict
- Description
JSON array with the input
records, each record enriched by field_objects(of type Slab Label or Card). The main object contains_identificationwith necessary information of identified card.
- Name
status- Type
- dict
- Description
A JSON map/dictionary with a status of the method processing. It contains these subfields:
code(numeric code of the operation status; it follows the concept of HTTP status codes) andtext(text describing the status code).
Request
curl https://api.ximilar.com/collectibles/v2/tcg_id -H "Content-Type: application/json" -H "Authorization: Token __API_TOKEN__" -d '{
"records": [
{
"_url": "__PATH_TO_IMAGE_URL__"
}
],
"slab_grade": true,
"slab_id": true,
"pricing": true
}'
Response
{
"records": [
{
"_url": "https://images.ximilar.com/examples/demo/collectibles/pokemon.jpeg",
"_status": { "code": 200, "text": "OK", "request_id": "80985c61-bc9b-49f2-9756-c09c926b015c"},
"_id": "c945fbd1-b706-4e83-a4f2-4cc9ea967e1d",
"_width": 389,
"_height": 659,
"Category": "Card/Trading Card Game",
"_objects": [
{
"name": "Card",
"id": "76fa9ec3-e0d9-408a-b582-55a1cd6712e0",
"bound_box": [
38,
165,
348,
616
],
"prob": 0.9330817461013794,
"area": 0.5453850384823933,
"Top Category": [
{
"id": "8ae26c4a-ae79-4c01-9b54-ac4e2b42e914",
"name": "Card",
"prob": 1.0
}
],
"_tags": {
"Category": [
{
"name": "Card/Trading Card Game",
"prob": 1.0,
"id": "d44efa56-09c3-4829-b3e7-226fe9cfe798",
"pre-filled": true
}
],
"Side": [
{
"prob": 0.99487,
"name": "front",
"id": "b48a23fe-381e-4a67-b649-9839e1f7d5a7"
}
],
"Subcategory": [
{
"prob": 0.99945,
"name": "Pokemon",
"id": "fdbb3942-b7e0-476a-8f13-af77aedd13e6"
}
],
"Foil/Holo": [
{
"prob": 0.99159,
"name": "Foil/Holo",
"id": "f8ade5be-18f3-49d2-aa76-2431a16bd3f7"
}
],
"Alphabet": [
{
"prob": 0.99615,
"name": "latin",
"id": "f0339837-aa0b-414f-acb8-d47d6ef3b7fa"
}
]
},
"_tags_simple": [
"Card/Trading Card Game",
"front",
"Pokemon",
"Foil/Holo",
"latin"
],
"_identification": {
"best_match": {
"year": 2000,
"full_name": "Dark Gyarados Team Rocket (RO) #8",
"name": "Dark Gyarados",
"set": "Team Rocket",
"color": "Water",
"type": "Pokemon",
"set_code": "RO",
"rarity": "Rare Holo",
"out_of": "82",
"card_number": "8",
"series": "Classic",
"subcategory": "Pokemon",
"links": {
"tcgplayer.com": "https://www.tcgplayer.com/product/84606",
"ebay.com": "https://www.ebay.com/sch/i.html?_nkw=Pokemon+Dark+Gyarados+RO+%238&_sacat=2536"
},
"pricing": {
"list": [
{
"item_id": "v1|116472483717|0",
"item_link": "https://www.ebay.com/itm/116472483717",
"name": "Dark Gyarados 8/82 Holo Team Rocket Prerelease Promo Pokemon Card -",
"price": 11.99,
"currency": "USD",
"country_code": "US",
"source": "eBay",
"date_of_creation": "2025-02-09",
"date_of_sale": null,
"grade_company": null,
"grade": null,
"version": "Foil/Holo",
"variation": null
},
{
"item_id": "v1|335803065051|0",
"item_link": "https://www.ebay.com/itm/335803065051",
"name": "2000 Pokemon Team Rocket Holo Dark Gyarados #8 0n5m",
"price": 40.5,
"currency": "USD",
"country_code": "US",
"source": "eBay",
"date_of_creation": "2025-02-08",
"date_of_sale": null,
"grade_company": null,
"grade": null,
"version": "Foil/Holo",
"variation": null
}
]
}
},
"alternatives": [
{
"year": 2000,
"full_name": "Dark Gyarados Team Rocket (RO) #25",
"name": "Dark Gyarados",
"set": "Team Rocket",
"color": "Water",
"type": "Pokemon",
"set_code": "RO",
"rarity": "Rare",
"out_of": "82",
"card_number": "25",
"series": "Classic",
"subcategory": "Pokemon",
"links": {
"tcgplayer.com": "https://www.tcgplayer.com/product/84607",
"ebay.com": "https://www.ebay.com/sch/i.html?_nkw=Pokemon+Dark+Gyarados+RO+%2325&_sacat=2536"
}
},
{
"year": 2021,
"full_name": "Dark Gyarados Celebrations (CEL) #8",
"name": "Dark Gyarados",
"set": "Celebrations",
"color": "Water",
"type": "Pokemon",
"set_series_code": "Ann25thR",
"set_code": "CEL",
"rarity": "Ultra Rare",
"out_of": "82",
"card_number": "8",
"series": "Sword & Shield",
"subcategory": "Pokemon",
"links": {
"tcgplayer.com": "https://www.tcgplayer.com/product/250294",
"ebay.com": "https://www.ebay.com/sch/i.html?_nkw=Pokemon+Dark+Gyarados+CEL+%238&_sacat=2536"
}
}
],
"distances": [
0.30719548,
0.38173488,
0.47646382
]
}
},
{
"name": "Slab Label",
"id": "9242aeda-1a35-4c93-b75c-958af91e63a7",
"bound_box": [
31,
22,
375,
130
],
"prob": 0.763236403465271,
"area": 0.14492629246618893,
"Top Category": [
{
"id": "79425eda-c25b-48e3-a9de-0fe8303a4635",
"name": "Slab Label",
"prob": 1.0
}
],
"_tags": {
"Company": [
{
"prob": 0.98724,
"name": "PSA",
"id": "2f4545e7-08e4-492b-ab4e-02a409d58988"
}
],
"Grade": [
{
"prob": 0.99235,
"name": "9",
"id": "5ae512e1-5296-4900-ac6d-3092b71d28e7"
}
],
"Side": [
{
"prob": 0.97328,
"name": "front",
"id": "6b41ebfc-cfbd-4f93-a9bf-bf54b357cafd"
}
],
"Graded": [
{
"prob": 0.9806,
"name": "yes",
"id": "9d7bf709-dfbc-4595-91be-a9d779b5f33c"
}
]
},
"_tags_simple": [
"PSA",
"9",
"front"
],
"_identification": {
"best_match": {
"name": "DARK GYARADOS",
"brand": "POKEMON",
"verbal_grade": "Mint",
"grade": "9",
"year": null,
"card_no": "#8",
"certificate_number": "48658983",
"lang": null,
"set": "ROCKET",
"rarity": "HOLO",
"links": {
"psacard.com": "https://www.psacard.com/cert/48658983"
}
}
}
}
]
}
],
"pricing": true,
"slab_grade": true,
"slab_id": true,
"status": {
"code": 200,
"text": "OK",
"request_id": "80985c61-bc9b-49f2-9756-c09c926b015c",
"proc_id": "a89e599c-d0cd-4e44-9202-e7a3634adb63"
},
"statistics": {
"processing time": 1.1655206680297852
}
}
Make the prediction faster and more precise
The endpoint can be faster and more accurate if you send Top Category, Category, Side, Alphabet, Subcategory, Foil/Holo and Rotation in the record. The more information you can send the faster and more precise the system is. You can even specify the set_code of the game, in this way the results are restricted on specific game.
{
"_url": "https://images.ximilar.com/examples/demo/collectibles/gideon.jpeg",
"Top Category": "Card",
"Side": "front",
"Alphabet": "latin",
"Category": "Card/Trading Card Game",
"Subcategory": "Magic The Gathering",
"Foil/Holo": "Non-Foil",
"Rotation": "rotation_ok",
"set_code": "CMM"
}
Sport Card Identification
Given a list of image records, this method returns identification for largest Sport Card detected on image.
This service can identify cards from various sports including Baseball, Basketball, Football, Hockey, Soccer, and MMA. It can also recognize visual features like Foil/Holo and Autograph. The service provides details such as name, team, year, card_number, set_name, and company.
Because there are several millions of sports cards and thousands of them are released every month, our database of sport cards is limited and so is our identification. Our database contains several millions of cards and we are constantly working on expanding the database. If you find that we are missing some cards feel free to contact us at tech@ximilar.com.
Required attributes
- Name
records- Type
- dict
- Description
A batch of json records (max 10), one record is representation of an image and it's defined by
_urlor_base64.
Optional attributes
- Name
slab_id- Type
- boolean
- Default
- Default:False
- Description
If set to true, then the slab identification will be done via OCR and AI, default False. This will analyse only one slab label on the image and also count additional API credits according to /v2/slab_id.
- Name
slab_grade- Type
- boolean
- Default
- Default:False
- Description
If set to true, then AI system will analyse the company and grade on the image, default False. This will also count additional API credits according to /v2/slab_grade.
- Name
analyze_all- Type
- boolean
- Default
- Default:False
- Description
If specified it will analyze all cards on image (only 1 record/image is allowed), the perfect use case is a card binder. The number of credits is per number of analysed cards on image instead of records (cards count * credit price of endpoint).
- Name
pricing- Type
- boolean
- Default
- Default:False
- Description
If set to true, then system will try to return the prices, default False. This will also count additional API credits according to pricing operation. The API credits will not be counted if the system return no prices.
Returns
HTTP error code 2XX, if the method was OK and other HTTP error code, if the method failed. Body of the response is a JSON object (map) with the following fields:
- Name
records- Type
- dict
- Description
JSON array with the input
records, each record enriched by field_objects(of type Slab Label or Card). The main object contains_identificationwith necessary information of identified card.
- Name
status- Type
- dict
- Description
A JSON map/dictionary with a status of the method processing. It contains these subfields:
code(numeric code of the operation status; it follows the concept of HTTP status codes) andtext(text describing the status code).
Request
curl https://api.ximilar.com/collectibles/v2/sport_id -H "Content-Type: application/json" -H "Authorization: Token __API_TOKEN__" -d '{
"records": [
{
"_url": "__PATH_TO_IMAGE_URL__"
}
],
"pricing": true
}'
Response
{
"records": [
{
"_url": "https://images.ximilar.com/examples/demo/collectibles/jirip.jpeg",
"_status": {
"code": 200,
"text": "OK",
"request_id": "cf82301a-46a5-4648-8909-a4894ece97c6"
},
"_id": "0aa7edf6-2495-4d0e-8502-f8dd7da8e8c7",
"_width": 1200,
"_height": 1600,
"Category": "Card/Sport Card",
"_objects": [
{
"name": "Card",
"id": "76fa9ec3-e0d9-408a-b582-55a1cd6712e0",
"bound_box": [
77,
97,
1096,
1464
],
"prob": 0.8502776026725769,
"area": 0.7255067708333334,
"Top Category": [
{
"id": "8ae26c4a-ae79-4c01-9b54-ac4e2b42e914",
"name": "Card",
"prob": 1.0
}
],
"_tags": {
"Category": [
{
"name": "Card/Sport Card",
"prob": 1.0,
"id": "a5634621-4a37-4b37-aa3d-720b2d6b35ec",
"pre-filled": true
}
],
"Side": [
{
"prob": 0.97608,
"name": "front",
"id": "651c8141-2b18-479b-a8b1-b959bc34b729"
}
],
"Subcategory": [
{
"prob": 0.84972,
"name": "MMA",
"id": "87e0dd9b-68f2-42f2-9283-2379c43f790b"
}
],
"Autograph": [
{
"prob": 0.97988,
"name": "not signed",
"id": "5c1bb98b-cd69-42c9-ab2a-75e480ffa2b0"
}
],
"Foil/Holo": [
{
"prob": 0.99476,
"name": "Foil/Holo",
"id": "8d0a5c67-eb08-4908-81c9-60e74d7028e7"
}
],
"Graded": [
{
"prob": 0.97944,
"name": "no",
"id": "3a532911-7d7c-4b9b-8442-f0f293952be6"
}
]
},
"_tags_simple": [
"Card/Sport Card",
"front",
"MMA",
"not signed",
"Foil/Holo"
],
"_identification": {
"best_match": {
"year": "2021",
"name": "Jiri Prochazka",
"set_name": "Chronicles",
"card_type": "Rookie Card",
"card_number": "128",
"subcategory": "MMA",
"sub_set": "UFC",
"company": "Panini",
"full_name": "Jiri Prochazka 2021 #128 Panini Chronicles UFC",
"links": {
"ebay.com": "https://www.ebay.com/sch/i.html?_nkw=Jiri+Prochazka+2021+Chronicles+UFC+%23128+RC&_sacat=212",
"comc.com": "https://www.comc.com/Cards,=Jiri+Prochazka+2021+Chronicles+UFC+%23128+RC",
"beckett.com": "https://marketplace.beckett.com/search_new/?term=Jiri+Prochazka+2021+Chronicles+UFC+%23128+RC"
},
"pricing": {
"list": [
{
"item_id": "v1|167302788807|0",
"item_link": "https://www.ebay.com/itm/167302788807",
"name": "SILVER RC Jiri Prochazka 2021 Panini Chronicles UFC Certified RC #128 MMA 📈",
"price": 5.0,
"currency": "USD",
"country_code": "US",
"source": "eBay",
"date_of_creation": "2025-02-08",
"date_of_sale": null,
"grade_company": null,
"grade": null,
"version": "Foil/Holo",
"variation": null
}
]
}
},
"alternatives": [
{
"year": "2021",
"name": "Jiri Prochazka",
"set_name": "Chronicles",
"card_type": "Rookie Card",
"card_number": "128",
"sub_set": "UFC",
"subcategory": "MMA",
"company": "Panini",
"full_name": "Jiri Prochazka 2021 #128 Panini Chronicles UFC",
"links": {
"ebay.com": "https://www.ebay.com/sch/i.html?_nkw=Jiri+Prochazka+2021+Chronicles+UFC+%23128+RC&_sacat=212",
"comc.com": "https://www.comc.com/Cards,=Jiri+Prochazka+2021+Chronicles+UFC+%23128+RC",
"beckett.com": "https://marketplace.beckett.com/search_new/?term=Jiri+Prochazka+2021+Chronicles+UFC+%23128+RC"
}
}
],
"distances": [
0.20778613,
0.33050093
]
}
}
],
"Graded Slab": [
{
"prob": 0.97944,
"name": "no",
"id": "3a532911-7d7c-4b9b-8442-f0f293952be6"
},
{
"prob": 0.02056,
"name": "yes",
"id": "9d7bf709-dfbc-4595-91be-a9d779b5f33c"
}
]
}
],
"pricing": true,
"status": {
"code": 200,
"text": "OK",
"request_id": "cf82301a-46a5-4648-8909-a4894ece97c6",
"proc_id": "3b7a5c94-2ebb-4586-ad72-859bae9ac8a8"
},
"statistics": {
"processing time": 1.959585428237915
}
}
Comics Identification
Given a list of images/records, this method returns identification for largest Comics detected on image.
This service can identify comics books, comics magazines and manga. This endpoint will tell you name of the person, title, date of release, issue number, publisher and origin_date. We believe that these few informations are enough to precisely identify the comics.
Because there are several millions of comics and hundreds of them are released every month, our database of comics is limited and so is our identification. We are constantly working on expanding the database. If you find that we are missing some comics feel free to contact us at tech@ximilar.com.
Required attributes
- Name
records- Type
- dict
- Description
A batch of json records (max 10), one record is representation of an image and it's defined by
_urlor_base64.
Optional attributes
- Name
slab_id- Type
- boolean
- Default
- Default:False
- Description
If set to true, then the slab identification will be done via OCR and AI, default False. This will analyse only one slab label on the image and also count additional API credits according to /v2/slab_id.
- Name
slab_grade- Type
- boolean
- Default
- Default:False
- Description
If set to true, then AI system will analyse the company and grade on the image, default False. This will also count additional API credits according to /v2/slab_grade.
- Name
analyze_all- Type
- boolean
- Default
- Default:False
- Description
If specified it will analyze all comics on image (only 1 record/image is allowed), the number of credits is per number of analysed comics on image instead of records (comics count * credit price of endpoint)
- Name
pricing- Type
- boolean
- Default
- Default:False
- Description
If set to true, then system will try to return the prices, default False. This will also count additional API credits according to pricing operation. The API credits will not be counted if the system return no prices.
Returns
HTTP error code 2XX, if the method was OK and other HTTP error code, if the method failed. Body of the response is a JSON object (map) with the following fields:
- Name
records- Type
- dict
- Description
JSON array with the input
records, each record enriched by field_objects(of type Slab Label or Comics). The main object contains_identificationwith necessary information of identified comics.
- Name
status- Type
- dict
- Description
A JSON map/dictionary with a status of the method processing. It contains these subfields:
code(numeric code of the operation status; it follows the concept of HTTP status codes) andtext(text describing the status code).
Request
curl https://api.ximilar.com/collectibles/v2/comics_id -H "Content-Type: application/json" -H "Authorization: Token __API_TOKEN__" -d '{
"records": [
{
"_url": "__PATH_TO_IMAGE_URL__"
}
],
"slab_id": true
}'
Response
{
"records": [
{
"_url": "__PATH_TO_IMAGE_URL__",
"_status": {
"code": 200,
"text": "OK",
"request_id": "d0bbbb58-490f-4721-8273-e32bc01c9bc6"
},
"_id": "dec258cf-4d77-4c55-b4b7-418725b4be08",
"_width": 419,
"_height": 640,
"_objects": [
{
"name": "Comics",
"id": "520eee19-2502-41a7-a1cf-23c6daf22138",
"bound_box": [
38,
93,
370,
621
],
"prob": 0.7870738506317139,
"area": 0.6536992840095466,
"Top Category": [
{
"id": "6abbefd9-5571-4915-a057-2fba3c3663a3",
"name": "Comics",
"prob": 1.0
}
],
"_identification": {
"best_match": {
"name": "Fantastic Four",
"title": "Counter-Earth Must Die--At The Hand Of Galactus!",
"date": "August 1976",
"number": "173",
"publisher": "Marvel",
"origin_date": "1961",
"links": {
"ebay.com": "https://www.ebay.com/sch/i.html?_nkw=Comics+Fantastic+Four+%23173+Marvel",
"midtowncomics.com": "https://www.midtowncomics.com/search?q=Comics+Fantastic+Four+%23173+Marvel"
}
},
"alternatives": []
}
},
{
"name": "Slab Label",
"id": "9242aeda-1a35-4c93-b75c-958af91e63a7",
"bound_box": [
20,
14,
375,
101
],
"prob": 0.583219051361084,
"area": 0.115173776849642,
"Top Category": [
{
"id": "79425eda-c25b-48e3-a9de-0fe8303a4635",
"name": "Slab Label",
"prob": 1.0
}
],
"_tags": {
"Category": [
{
"prob": 0.73168,
"name": "CGC-COMICS",
"id": "b2aae47e-a788-46c7-98f5-b27d3d1bdd3c"
}
],
"Side": [
{
"prob": 0.94687,
"name": "front",
"id": "6b41ebfc-cfbd-4f93-a9bf-bf54b357cafd"
}
],
"Graded": [
{
"prob": 0.95005,
"name": "yes",
"id": "9d7bf709-dfbc-4595-91be-a9d779b5f33c"
}
]
},
"_tags_simple": [
"CGC-COMICS",
"front"
],
"_ocr": {
"full_text": "CGC UNIVERSAL GRADE 7.5 Fantastic Four173 Marvel Comic.1/76",
"lang_name": "english",
"lang": "en"
},
"_identification": {
"best_match": {
"name": "FANTASTIC FOUR",
"brand": "MARVEL COMIC",
"verbal_grade": "Fantastic",
"grade": "7.5",
"year": "1976",
"card_no": "173",
"certificate_number": null,
"lang": null
}
}
}
],
"Graded Slab": [
{
"prob": 0.95005,
"name": "yes",
"id": "9d7bf709-dfbc-4595-91be-a9d779b5f33c"
},
{
"prob": 0.04995,
"name": "no",
"id": "3a532911-7d7c-4b9b-8442-f0f293952be6"
}
]
}
],
"slab_id": true,
"status": {
"code": 200,
"text": "OK",
"request_id": "d0bbbb58-490f-4721-8273-e32bc01c9bc6",
"proc_id": "cfc7eff0-05a1-47e4-8e30-79bad9253dd5"
},
"statistics": {
"processing time": 0.7930326461791992
}
}
Card OCR & Identification
Given a list of image records, this method returns and identification for largest card detected (and largest slab label if present) on image. This method is suitable for reading text Cards and Slab Labels (PSA, BECKETT, ...). The identification is done by mining of information from ocr text via AI model (and that is why is sometimes not exactly precise).
Partially correct identification (missing year) for card can look like this:
{
"name": "ERLING BRAUT HAALAND",
"year": null,
"subcategory": "soccer"
}
Correct identification for graded slab label can look like this:
{
"name": "2019 FINEST UEFA ERLING BRAUT HAALAND",
"grade": "10",
"year": "2019",
"certificate_number": "51974133"
}
Required attributes
- Name
records- Type
- dict
- Description
A batch of json records (max 10), one record is representation of an image and it's defined by
_urlor_base64.
Returns
HTTP error code 2XX, if the method was OK and other HTTP error code, if the method failed. Body of the response is a JSON object (map) with the following fields:
- Name
records- Type
- dict
- Description
JSON array with the input
records, each record enriched by field_objects(of type Slab Label or Card)._objects(Card,Slab Label) contains_ocr,_tags,_tags_simpleand_identification. The_identificationcontainsbest_matchfield. The identification returns info likename,full_name,year,grade,certificate_number...
- Name
status- Type
- dict
- Description
A JSON map/dictionary with a status of the method processing. It contains these subfields:
code(numeric code of the operation status; it follows the concept of HTTP status codes) andtext(text describing the status code).
Request
curl https://api.ximilar.com/collectibles/v2/card_ocr_id -H "Content-Type: application/json" -H "Authorization: Token __API_TOKEN__" -d '{
"records": [
{
"_url": "__PATH_TO_IMAGE_URL__"
}
]
}'
Response
{
"records": [
{
"_url": "__URL_PATH__",
"_status": {
"code": 200,
"text": "OK",
"request_id": "37d5842e-a6f1-443f-a2c4-6cafa5ed604b"
},
"_id": "8c26e90a-627e-444e-a3a6-92845f581858",
"_objects": [
{
"name": "Card",
"id": "76fa9ec3-e0d9-408a-b582-55a1cd6712e0",
"bound_box": [
213,
403,
872,
1273
],
"prob": 0.978815495967865,
"area": 0.36865354938271605,
"Top Category": [
{
"id": "8ae26c4a-ae79-4c01-9b54-ac4e2b42e914",
"name": "Card",
"prob": 1.0
}
],
"_tags": {
"Category": [
{
"prob": 0.99694,
"name": "Card/Sport Card",
"id": "a5634621-4a37-4b37-aa3d-720b2d6b35ec"
}
],
"Side": [
{
"prob": 0.98772,
"name": "front",
"id": "651c8141-2b18-479b-a8b1-b959bc34b729"
}
],
"Subcategory": [
{
"prob": 0.96789,
"name": "soccer",
"id": "3275f70c-10f4-4807-89ca-f166e8c3aa7c"
}
],
"Autograph": [
{
"prob": 0.97277,
"name": "not autograph",
"id": "5c1bb98b-cd69-42c9-ab2a-75e480ffa2b0"
}
]
},
"_tags_simple": [
"Card/Sport Card",
"front",
"soccer",
"not autograph"
],
"_ocr": {
"texts": [
{
"polygon": [
[
61.0,
60.0
],
[
159.0,
55.0
],
[
160.0,
76.0
],
[
61.0,
81.0
]
],
"text": "Tirest",
"prob": 0.8078048825263977
},
{
"polygon": [
[
519.0,
97.0
],
[
557.0,
97.0
],
[
557.0,
111.0
],
[
519.0,
111.0
]
],
"text": "SEFA",
"prob": 0.7368016839027405
},
{
"polygon": [
[
481.0,
113.0
],
[
597.0,
111.0
],
[
597.0,
129.0
],
[
482.0,
131.0
]
],
"text": "CHAMPIONS",
"prob": 0.9519932866096497
},
{
"polygon": [
[
508.0,
127.0
],
[
573.0,
129.0
],
[
572.0,
146.0
],
[
508.0,
144.0
]
],
"text": "LEAGUE,",
"prob": 0.8668462634086609
},
{
"polygon": [
[
207.0,
300.0
],
[
372.0,
307.0
],
[
370.0,
363.0
],
[
205.0,
356.0
]
],
"text": "Red Bul",
"prob": 0.9309224486351013
},
{
"polygon": [
[
61.0,
712.0
],
[
250.0,
684.0
],
[
254.0,
719.0
],
[
66.0,
747.0
]
],
"text": "ERLING BRAUT",
"prob": 0.9668567180633545
},
{
"polygon": [
[
186.0,
753.0
],
[
370.0,
730.0
],
[
375.0,
773.0
],
[
191.0,
795.0
]
],
"text": "HAALAND",
"prob": 0.9960637092590332
},
{
"polygon": [
[
330.0,
786.0
],
[
471.0,
766.0
],
[
474.0,
788.0
],
[
333.0,
807.0
]
],
"text": "FC SALZBURG",
"prob": 0.9561893343925476
}
],
"full_text": "Tirest SEFA CHAMPIONS LEAGUE, Red Bul ERLING BRAUT HAALAND FC SALZBURG ",
"lang_name": "german",
"lang": "de"
},
"_identification": {
"best_match": {
"name": "ERLING BRAUT HAALAND",
"year": null,
"subcategory": "soccer"
}
}
},
{
"name": "Slab Label",
"id": "9242aeda-1a35-4c93-b75c-958af91e63a7",
"bound_box": [
205,
108,
868,
309
],
"prob": 0.7645022869110107,
"area": 0.0856886574074074,
"_tags": {
"Category": [
{
"prob": 0.97388,
"name": "PSA",
"id": "2f4545e7-08e4-492b-ab4e-02a409d58988"
}
]
},
"_tags_simple": [
"PSA"
],
"_ocr": {
"full_text": "#91 2019 FINESTUCL ERLING BRAUT HAALAND GEMMT 10 51974133 ",
"lang_name": "german",
"lang": "de"
},
"_identification": {
"best_match": {
"name": "2019 FINEST UEFA ERLING BRAUT HAALAND",
"grade": "10",
"year": "2019",
"certificate_number": "51974133"
}
}
}
]
}
],
"statistics": {
"processing time": 6.20737099647522
},
"status": {
"code": 200,
"text": "OK",
"request_id": "37d5842e-a6f1-443f-a2c4-6cafa5ed604b",
"proc_id": "189995ef-c1e0-4799-a35e-ad32f4730c81"
}
}
Slab ID
Given a list of image records, this method returns detected objects (Card, Slab Label) with OCR analysis. The result Slab Label object with _objects (_tags and _ocr included), which represents location of the detected items (and read text). The Slab Label contains information about company/category (BECKETT, CGC, PSA, SGC, ACE, MANA, TAG, OTHER). For one Slab Label per image, this model analyzes read text via AI of the slab and returns the analyzed info in _identification field.
If you would like to analyse labels from company which we do not include, feel free to contact us at tech@ximilar.com
Required attributes
- Name
records- Type
- dict
- Description
A batch of json records (max 10), one record is representation of an image and it's defined by
_urlor_base64.
Returns
HTTP error code 2XX, if the method was OK and other HTTP error code, if the method failed. Body of the response is a JSON object (map) with the following fields:
- Name
records- Type
- dict
- Description
JSON array with the input
records, each record enriched by fields,Slab Labeland_objectswith_tagsand_ocr.
- Name
status- Type
- dict
- Description
A JSON map/dictionary with a status of the method processing. It contains these subfields:
code(numeric code of the operation status; it follows the concept of HTTP status codes) andtext(text describing the status code).
Request
curl https://api.ximilar.com/collectibles/v2/slab_id -H "Content-Type: application/json" -H "Authorization: Token __API_TOKEN__" -d '{
"records": [
{
"_url": "__PATH_TO_IMAGE_URL__"
}
]
}'
Response
{
"records": [
{
"_url": "https://images.ximilar.com/examples/demo/collectibles/pokemon.jpeg",
"_status": {
"code": 200,
"text": "OK",
"request_id": "7f10ec20-262e-4a05-93ce-399e0813ef1d"
},
"_id": "51750739-c7a3-4ad7-aa0f-28d9fa27d845",
"_objects": [
{
"name": "Slab Label",
"id": "9242aeda-1a35-4c93-b75c-958af91e63a7",
"bound_box": [
88,
22,
361,
131
],
"prob": 0.6562235951423645,
"area": 0.11607912588599226,
"Top Category": [
{
"id": "79425eda-c25b-48e3-a9de-0fe8303a4635",
"name": "Slab Label",
"prob": 1.0
}
],
"_tags": {
"Company": [
{
"prob": 0.9808,
"name": "PSA",
"id": "2f4545e7-08e4-492b-ab4e-02a409d58988"
}
],
"Side": [
{
"prob": 0.94674,
"name": "front",
"id": "6b41ebfc-cfbd-4f93-a9bf-bf54b357cafd"
}
],
"Category": [
{
"prob": 0.9808,
"name": "Slab Label/PSA",
"id": "2f4545e7-08e4-492b-ab4e-02a409d58988"
}
]
},
"_tags_simple": [
"PSA",
"front"
],
"_ocr": {
"texts": [
{
"polygon": [
[
236.0,
20.0
],
[
263.0,
20.0
],
[
263.0,
45.0
],
[
236.0,
45.0
]
],
"text": "#8",
"prob": 0.989625096321106
},
{
"polygon": [
[
4.0,
24.0
],
[
159.0,
24.0
],
[
159.0,
43.0
],
[
4.0,
43.0
]
],
"text": "POKEMON ROCKET",
"prob": 0.9793843030929565
},
{
"polygon": [
[
1.0,
42.0
],
[
167.0,
42.0
],
[
167.0,
61.0
],
[
1.0,
61.0
]
],
"text": "GYARADOS-HOLO",
"prob": 0.9837774038314819
},
{
"polygon": [
[
216.0,
41.0
],
[
263.0,
41.0
],
[
263.0,
62.0
],
[
216.0,
62.0
]
],
"text": "MINT",
"prob": 0.9965052604675293
},
{
"polygon": [
[
246.0,
59.0
],
[
263.0,
59.0
],
[
263.0,
81.0
],
[
246.0,
81.0
]
],
"text": "9",
"prob": 0.9890574812889099
},
{
"polygon": [
[
84.0,
73.0
],
[
131.0,
73.0
],
[
131.0,
102.0
],
[
84.0,
102.0
]
],
"text": "PA",
"prob": 0.9283953905105591
},
{
"polygon": [
[
178.0,
76.0
],
[
264.0,
76.0
],
[
264.0,
97.0
],
[
178.0,
97.0
]
],
"text": "48658983",
"prob": 0.9988992810249329
}
],
"full_text": "#8 POKEMON ROCKET GYARADOS-HOLO MINT 9 PA 48658983",
"lang_name": "english",
"lang": "en"
},
"_identification": {
"best_match": {
"name": "GYARADOS",
"brand": "POKEMON",
"verbal_grade": "MINT",
"grade": "9",
"year": null,
"card_no": "#8",
"certificate_number": "48658983",
"lang": null,
"set": "ROCKET",
"rarity": "HOLO",
"links": {
"psacard.com": "https://www.psacard.com/cert/48658983"
}
}
}
},
{
"name": "Card",
"id": "76fa9ec3-e0d9-408a-b582-55a1cd6712e0",
"bound_box": [
37,
164,
353,
618
],
"prob": 0.9187777042388916,
"area": 0.559638932557314,
"Top Category": [
{
"id": "8ae26c4a-ae79-4c01-9b54-ac4e2b42e914",
"name": "Card",
"prob": 1.0
}
]
}
],
"_width": 389,
"_height": 659
}
],
"status": {
"code": 200,
"text": "OK",
"request_id": "7f10ec20-262e-4a05-93ce-399e0813ef1d",
"proc_id": "125f4de8-94fc-46c2-a67b-7482b253a575"
},
"statistics": {
"processing time": 1.936791181564331
}
}
Slab Grade
Given a list of image records, this method returns detected objects (Card, Slab Label) with image recognition analysis. The result Slab Label object with _objects (_tags included), which represents location of the detected items (and analysed features - Company, Grade, Side). The Slab Label object contains information about Company/Category (BECKETT, CGC, PSA, SGC, MANA, ACE, TAG, Other), Grade and Side.
If you would like to analyse labels from company which we do not include, feel free to contact us at tech@ximilar.com
Required attributes
- Name
records- Type
- dict
- Description
A batch of json records (max 10), one record is representation of an image and it's defined by
_urlor_base64.
Returns
HTTP error code 2XX, if the method was OK and other HTTP error code, if the method failed. Body of the response is a JSON object (map) with the following fields:
- Name
records- Type
- dict
- Description
JSON array with the input
records, each record enriched by fields,Slab Labeland_objectswith_tags.
- Name
status- Type
- dict
- Description
A JSON map/dictionary with a status of the method processing. It contains these subfields:
code(numeric code of the operation status; it follows the concept of HTTP status codes) andtext(text describing the status code).
Request
curl https://api.ximilar.com/collectibles/v2/slab_grade -H "Content-Type: application/json" -H "Authorization: Token __API_TOKEN__" -d '{
"records": [
{
"_url": "__PATH_TO_IMAGE_URL__"
}
]
}'
Response
{
"records": [
{
"_url": "https://images.ximilar.com/examples/demo/collectibles/pokemon.jpeg",
"_status": {
"code": 200,
"text": "OK",
"request_id": "ac414766-83e1-41f2-a589-7958c1fa9f23"
},
"_id": "1ee48bb5-e92f-44e7-90cd-72eb898b537e",
"_objects": [
{
"name": "Slab Label",
"id": "9242aeda-1a35-4c93-b75c-958af91e63a7",
"bound_box": [
88,
22,
361,
131
],
"prob": 0.6562235951423645,
"area": 0.11607912588599226,
"Top Category": [
{
"id": "79425eda-c25b-48e3-a9de-0fe8303a4635",
"name": "Slab Label",
"prob": 1.0
}
],
"_tags": {
"Company": [
{
"prob": 0.9808,
"name": "PSA",
"id": "2f4545e7-08e4-492b-ab4e-02a409d58988"
}
],
"Grade": [
{
"prob": 0.5204,
"name": "9",
"id": "445d621b-60c1-4e56-b33e-b82d2905ed9c"
}
],
"Side": [
{
"prob": 0.94674,
"name": "front",
"id": "6b41ebfc-cfbd-4f93-a9bf-bf54b357cafd"
}
]
},
"_tags_simple": [
"PSA",
"9",
"front"
]
},
{
"name": "Card",
"id": "76fa9ec3-e0d9-408a-b582-55a1cd6712e0",
"bound_box": [
37,
164,
353,
618
],
"prob": 0.9187777042388916,
"area": 0.559638932557314,
"Top Category": [
{
"id": "8ae26c4a-ae79-4c01-9b54-ac4e2b42e914",
"name": "Card",
"prob": 1.0
}
]
}
],
"Graded Slab": [
{
"prob": 0.9806,
"name": "yes",
"id": "9d7bf709-dfbc-4595-91be-a9d779b5f33c"
},
{
"prob": 0.0194,
"name": "no",
"id": "3a532911-7d7c-4b9b-8442-f0f293952be6"
}
],
"_width": 389,
"_height": 659
}
],
"status": {
"code": 200,
"text": "OK",
"request_id": "ac414766-83e1-41f2-a589-7958c1fa9f23",
"proc_id": "39e6072b-8430-4821-9047-1fecbf807930"
},
"statistics": {
"processing time": 0.8551025390625
}
}
Collectibles Detection
Given a list of image records, this method returns objects of collector's items for each of them. The result images with _objects which represents location of the detected items. This method can be used as a first step for other endpoints as /v2/tcg_id if you know that multiple cards are on the image.
Required attributes
- Name
records- Type
- dict
- Description
A batch of json records (max 10), one record is representation of an image and it's defined by
_urlor_base64.
Returns
HTTP error code 2XX, if the method was OK and other HTTP error code, if the method failed. Body of the response is a JSON object (map) with the following fields:
- Name
records- Type
- dict
- Description
JSON array with the input
records, each record enriched by fields (_objects).
- Name
status- Type
- dict
- Description
A JSON map/dictionary with a status of the method processing. It contains these subfields:
code(numeric code of the operation status; it follows the concept of HTTP status codes) andtext(text describing the status code).
Request
curl https://api.ximilar.com/collectibles/v2/detect -H "Content-Type: application/json" -H "Authorization: Token __API_TOKEN__" -d '{
"records": [
{
"_url": "__PATH_TO_IMAGE_URL__"
}
]
}'
Response
{
"records": [
{
"_url": "__IMAGE_URL__",
"_status": {
"code": 200,
"text": "OK",
"request_id": "1e459b5d-8bed-4d9a-8647-fc1c99997875"
},
"_id": "ebb627dd-e859-4677-83cf-612032d2ebb9",
"_objects": [
{
"name": "Card",
"id": "76fa9ec3-e0d9-408a-b582-55a1cd6712e0",
"bound_box": [
105,
348,
853,
1436
],
"prob": 0.9431861639022827,
"area": 0.5242204881844521,
"expand_by_bound_box": [
91,
327,
867,
1457
],
"Top Category": [
{
"id": "8ae26c4a-ae79-4c01-9b54-ac4e2b42e914",
"name": "Card",
"prob": 1.0
}
]
},
{
"name": "Slab Label",
"id": "9242aeda-1a35-4c93-b75c-958af91e63a7",
"bound_box": [
77,
32,
943,
249
],
"prob": 0.4136030375957489,
"area": 0.12104897690483277,
"expand_by_bound_box": [
60,
28,
960,
253
],
"Top Category": [
{
"id": "79425eda-c25b-48e3-a9de-0fe8303a4635",
"name": "Slab Label",
"prob": 1.0
}
]
}
],
"_width": 999,
"_height": 1554
}
],
"status": {
"code": 200,
"text": "OK",
"request_id": "1e459b5d-8bed-4d9a-8647-fc1c99997875",
"proc_id": "219ad5f2-ad11-440c-b337-4f78f304e47b"
},
"statistics": {
"processing time": 2.1185216903686523
}
}