Datasets:
image imagewidth (px) 447 5.71k | stem stringlengths 7 13 | tier stringclasses 3
values | doc_type stringclasses 17
values | gt_text stringlengths 45 16.1k | gt_json stringlengths 331 26.4k ⌀ | structured_gt stringlengths 1.08k 15.8k ⌀ | markdown_gt stringclasses 17
values | geometry_gt stringclasses 17
values |
|---|---|---|---|---|---|---|---|---|
invoice_000 | easy | invoice | Invoice no: 97159829 Date of issue: 09/18/2015 Seller: Client: Bradley-Andrade Castro PLC 9879 Elizabeth Common Unit 9678 Box 9664 Lake Jonathan, RI 12335 DPO AP 69387 Tax Id: 985-73-8194 Tax Id: 994-72-1270 IBAN: GB81LZWO32519172531418 ITEMS UM No. Description Qty Net price Net worth VAT [%] Gross worth 2,00 444,60 88... | {
"plain_text": "Invoice no: 97159829 Date of issue: 09/18/2015 Seller: Client: Bradley-Andrade Castro PLC 9879 Elizabeth Common Unit 9678 Box 9664 Lake Jonathan, RI 12335 DPO AP 69387 Tax Id: 985-73-8194 Tax Id: 994-72-1270 IBAN: GB81LZWO32519172531418 ITEMS UM No. Description Qty Net price Net worth VAT [%] Gross w... | {
"dataset": "mychen76/invoices-and-receipts_ocr_v1",
"split": "test",
"manifest_index": 0,
"dataset_offset": 0,
"doc_type": "invoice",
"json_schema": {
"type": "object",
"properties": {
"header": {
"type": "object",
"properties": {
"invoice_no": {
"type":... | <!-- markdown GT transcribed from the page image. Labels, column headers and
section order are what the document prints; values are the dataset's.
One valid rendering — score structure and content, not formatting. -->
# Invoice no: 97159829
Date of issue: 09/18/2015
## Seller:
Bradley-Andrade
9879 Elizabe... | null | |
invoice_001 | easy | invoice | Invoice no: 62517865. Date of issue: 06/02/2015 Seller: Client: Trujillo-Hunt Lee and Sons 430 Mark Ferry Suite 495 8552 Karen Islands Maxside, DC 65686 East Roger, ID 40416 Tax Id: 992-71-8540 Tax Id: 929-86-0601 IBAN: GB73HCPH34959888432899 ITEMS No. Description Qty UM Net price Net worth VAT [%] Gross worth Care & R... | {
"plain_text": "Invoice no: 62517865. Date of issue: 06/02/2015 Seller: Client: Trujillo-Hunt Lee and Sons 430 Mark Ferry Suite 495 8552 Karen Islands Maxside, DC 65686 East Roger, ID 40416 Tax Id: 992-71-8540 Tax Id: 929-86-0601 IBAN: GB73HCPH34959888432899 ITEMS No. Description Qty UM Net price Net worth VAT [%] G... | {
"dataset": "mychen76/invoices-and-receipts_ocr_v1",
"split": "test",
"manifest_index": 2,
"dataset_offset": 2,
"doc_type": "invoice",
"json_schema": {
"type": "object",
"properties": {
"header": {
"type": "object",
"properties": {
"invoice_no": {
"type":... | <!-- markdown GT transcribed from the page image. Labels, column headers and
section order are what the document prints; values are the dataset's.
One valid rendering — score structure and content, not formatting. -->
# Invoice no: 62517865
Date of issue: 06/02/2015
## Seller:
Trujillo-Hunt
430 Mark Ferry... | null | |
invoice_002 | easy | invoice | Invoice no: 16662010 Date of issue: 08/28/2016 Seller: Client: Smith-Cook Snyder-Johnson 174 Justin Causeway 05173 Heather Mill West Michaelmouth, ME 69894 Jenniferfort, wv 79662 Tax Id: 959-84-2124 Tax Id: 938-85-4960 IBAN: GB20BAKH22085364527355 ITEMS Qty UM Net price VAT [%] No. Description Net worth Gross worth 65,... | {
"plain_text": "Invoice no: 16662010 Date of issue: 08/28/2016 Seller: Client: Smith-Cook Snyder-Johnson 174 Justin Causeway 05173 Heather Mill West Michaelmouth, ME 69894 Jenniferfort, wv 79662 Tax Id: 959-84-2124 Tax Id: 938-85-4960 IBAN: GB20BAKH22085364527355 ITEMS Qty UM Net price VAT [%] No. Description Net wo... | {
"dataset": "mychen76/invoices-and-receipts_ocr_v1",
"split": "test",
"manifest_index": 4,
"dataset_offset": 4,
"doc_type": "invoice",
"json_schema": {
"type": "object",
"properties": {
"header": {
"type": "object",
"properties": {
"invoice_no": {
"type":... | <!-- markdown GT transcribed from the page image. Labels, column headers and
section order are what the document prints; values are the dataset's.
One valid rendering — score structure and content, not formatting. -->
# Invoice no: 16662010
Date of issue: 08/28/2016
## Seller:
Smith-Cook
174 Justin Causew... | null | |
invoice_003 | easy | invoice | Invoice no: 26343874. Date of issue: 07/17/2013 Seller: Client: Smith Ltd Herring-Floyd 290 Jodi Gardens 8113 Hansen Cliff Apt. 826 Charlesview, TN 10958 Port Alice, ID 99663 Tax Id: 916-97-6743 Tax Id: 913-80-4636 IBAN: GB78TZMX55978564099007 ITEMS Qty UM Net price VAT [%] No. Description Net worth Gross worth 5,00 24... | {
"plain_text": "Invoice no: 26343874. Date of issue: 07/17/2013 Seller: Client: Smith Ltd Herring-Floyd 290 Jodi Gardens 8113 Hansen Cliff Apt. 826 Charlesview, TN 10958 Port Alice, ID 99663 Tax Id: 916-97-6743 Tax Id: 913-80-4636 IBAN: GB78TZMX55978564099007 ITEMS Qty UM Net price VAT [%] No. Description Net worth ... | {
"dataset": "mychen76/invoices-and-receipts_ocr_v1",
"split": "test",
"manifest_index": 6,
"dataset_offset": 6,
"doc_type": "invoice",
"json_schema": {
"type": "object",
"properties": {
"header": {
"type": "object",
"properties": {
"invoice_no": {
"type":... | <!-- markdown GT transcribed from the page image. Labels, column headers and
section order are what the document prints; values are the dataset's.
One valid rendering — score structure and content, not formatting. -->
# Invoice no: 26343874
Date of issue: 07/17/2013
## Seller:
Smith Ltd
290 Jodi Gardens
C... | null | |
invoice_004 | easy | invoice | Invoice no: 25698672 Date of issue:. 08/01/2015 Seller: Client: Gordon, Alvarado and Jones. Manning-Ho 943 Sean Park Suite 284 92026 Michael Terrace Apt. 138 North Derrick, OR 67536 West Michael, IL 42693 Tax Id: 936-79-2794 Tax Id: 960-85-1578 IBAN: GB47JKAT88391563723291 ITEMS No. Description Qty UM Net price Net wor... | {
"plain_text": "Invoice no: 25698672 Date of issue:. 08/01/2015 Seller: Client: Gordon, Alvarado and Jones. Manning-Ho 943 Sean Park Suite 284 92026 Michael Terrace Apt. 138 North Derrick, OR 67536 West Michael, IL 42693 Tax Id: 936-79-2794 Tax Id: 960-85-1578 IBAN: GB47JKAT88391563723291 ITEMS No. Description Qty U... | {
"dataset": "mychen76/invoices-and-receipts_ocr_v1",
"split": "test",
"manifest_index": 8,
"dataset_offset": 8,
"doc_type": "invoice",
"json_schema": {
"type": "object",
"properties": {
"header": {
"type": "object",
"properties": {
"invoice_no": {
"type":... | <!-- markdown GT transcribed from the page image. Labels, column headers and
section order are what the document prints; values are the dataset's.
One valid rendering — score structure and content, not formatting.
Item 3's description ends in CJK characters the dataset transcribed as
a run of zeros;... | null | |
invoice_005 | easy | invoice | Invoice no: 72455228 Date of issue: 05/11/2021 Seller: Client: Potts, Reed and Miller Brady LLC USCGC Weeks 26039 Anita Stravenue FPO AA 06868 North Peter, WA 22187 Tax Id: 913-82-5781 Tax Id: 953-75-0545 IBAN: GB54RDGK95400103726634 ITEMS No. Description Qty UM Net price Net worth VAT [%] Gross worth PlayStation 5 Con... | {
"plain_text": "Invoice no: 72455228 Date of issue: 05/11/2021 Seller: Client: Potts, Reed and Miller Brady LLC USCGC Weeks 26039 Anita Stravenue FPO AA 06868 North Peter, WA 22187 Tax Id: 913-82-5781 Tax Id: 953-75-0545 IBAN: GB54RDGK95400103726634 ITEMS No. Description Qty UM Net price Net worth VAT [%] Gross wort... | {
"dataset": "mychen76/invoices-and-receipts_ocr_v1",
"split": "test",
"manifest_index": 10,
"dataset_offset": 10,
"doc_type": "invoice",
"json_schema": {
"type": "object",
"properties": {
"header": {
"type": "object",
"properties": {
"invoice_no": {
"type... | <!-- markdown GT transcribed from the page image. Labels, column headers and
section order are what the document prints; values are the dataset's.
One valid rendering — score structure and content, not formatting.
Item 1's description ends in CJK characters the dataset transcribed as
"00OO"; kept as... | null | |
sroie_000 | easy | receipt | TAN CHAY YEE *** COPY *** OJC MARKETING SDN BHD ROC NO: 538358-H NO 2 & 4 BANDAR SERI ALAM 81750 MASAI TEL:07-388 2218 FAX:07-388 8218 EMAIL:NG@OJCGROUP.COM TAX INVOICE INVOICE NO : PEGIV-1030765 DATE : 15/01/2019 11:05:16 AM CASHIER : NG CHUAN MIN SALES PERSON : FATIN BILL TO : THE PEAK QUARRY WORKS ADDRESS :. DESCRIP... | {
"plain_text": "TAN CHAY YEE *** COPY *** OJC MARKETING SDN BHD ROC NO: 538358-H NO 2 & 4 BANDAR SERI ALAM 81750 MASAI TEL:07-388 2218 FAX:07-388 8218 EMAIL:NG@OJCGROUP.COM TAX INVOICE INVOICE NO : PEGIV-1030765 DATE : 15/01/2019 11:05:16 AM CASHIER : NG CHUAN MIN SALES PERSON : FATIN BILL TO : THE PEAK QUARRY WORKS... | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"manifest_index": 0,
"dataset_offset": 0,
"doc_type": "receipt",
"note": "One schema covering every extractable value on the receipt. The dataset ships only the four SROIE Task 3 targets (company, date, address, total) because that is the task it was b... | null | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"index": 0,
"image_size": [
463,
894
],
"normalisation": "pixel boxes divided by this shipped image's width/height",
"granularity": "line",
"verified": "39/40 sampled box texts found in the text GT",
"words": [
{
"text": "TAN CHAY YEE",
"bbox"... | |
sroie_001 | easy | receipt | PERNIAGAAN ZHENG HUI JM0326955-V NO.59 JALAN PERMAS 9/5 BANDAR BARU PERMAS JAYA 81750 JOHOR BAHRU TEL:07-386 7524 FAX:07-386 3793 GST NO: 000800689824 SIMPLIFIED TAX INVOICE GOGIANT ENGINEERING (M) SDN BHD RECEIPT#: CS00082258 SALESPERSON: DATE:09/02/2018 CASHIER: USER TIME:08:32:00 ITEM QTY (RM) RSP (RM) AMOUNT 6783 5... | {
"plain_text": "PERNIAGAAN ZHENG HUI JM0326955-V NO.59 JALAN PERMAS 9/5 BANDAR BARU PERMAS JAYA 81750 JOHOR BAHRU TEL:07-386 7524 FAX:07-386 3793 GST NO: 000800689824 SIMPLIFIED TAX INVOICE GOGIANT ENGINEERING (M) SDN BHD RECEIPT#: CS00082258 SALESPERSON: DATE:09/02/2018 CASHIER: USER TIME:08:32:00 ITEM QTY (RM) RSP... | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"manifest_index": 2,
"dataset_offset": 2,
"doc_type": "receipt",
"note": "Extended from the four SROIE Task 3 targets (company, date, address, total) to every extractable value on the page. The four-field schema is too thin to separate engines; the add... | null | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"index": 2,
"image_size": [
992,
1403
],
"normalisation": "pixel boxes divided by this shipped image's width/height",
"granularity": "line",
"verified": "40/40 sampled box texts found in the text GT",
"words": [
{
"text": "PERNIAGAAN ZHENG HUI",
... | |
sroie_002 | easy | receipt | SIN LIANHAP SDN BHD LOT 13 SIN LIANHAP SDN BHD LOT 13 KG BATU 30 44300 BTG KALI TEL:03-60752222(HUNTING LINE) FAX: 03-60752572 (COMPANY REG NO: 284922-D) (GST REG NO: 001610833920) TAX INVOICE CASH CUSTOMER INVOLCE NO.: :H0003939 DATE: :05/02/2018 CASHIER#: : RM CODE PPD 4MM DENLIME 1.887 SR 1.000PC 2.00 6023# GARDEN 5... | {
"plain_text": "SIN LIANHAP SDN BHD LOT 13 SIN LIANHAP SDN BHD LOT 13 KG BATU 30 44300 BTG KALI TEL:03-60752222(HUNTING LINE) FAX: 03-60752572 (COMPANY REG NO: 284922-D) (GST REG NO: 001610833920) TAX INVOICE CASH CUSTOMER INVOLCE NO.: :H0003939 DATE: :05/02/2018 CASHIER#: : RM CODE PPD 4MM DENLIME 1.887 SR 1.000PC ... | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"manifest_index": 5,
"dataset_offset": 5,
"doc_type": "receipt",
"note": "Extended from the four SROIE Task 3 targets (company, date, address, total) to every extractable value on this crumpled thermal receipt. Added the two registration numbers, the i... | null | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"index": 5,
"image_size": [
593,
1769
],
"normalisation": "pixel boxes divided by this shipped image's width/height",
"granularity": "line",
"verified": "39/40 sampled box texts found in the text GT",
"words": [
{
"text": "SIN LIANHAP SDN BHD",
... | |
sroie_003 | easy | receipt | SWC ENTERPRISE SDN BHD (1125830-V) NO. 5-7 SEKYSEN 4 BATANG KALI TEL : 03-6057 1377 TAX INVOICE (GST ID NO. : 002017808384) 08/01/2018 11:07:06 CASHIER: 123 NO:0100080332 ITEM/DESC. 20X30 BEG 1KG91X30) 20X30 1KG TOTAL QTY : 1 1 8.00 8.00 TOTAL QTY : TOTAL AMOUNT CASH CHANGE 8.00 10.00 2.00 G 0.45 THANK YOU ! PLEASE COM... | {
"plain_text": "SWC ENTERPRISE SDN BHD (1125830-V) NO. 5-7 SEKYSEN 4 BATANG KALI TEL : 03-6057 1377 TAX INVOICE (GST ID NO. : 002017808384) 08/01/2018 11:07:06 CASHIER: 123 NO:0100080332 ITEM/DESC. 20X30 BEG 1KG91X30) 20X30 1KG TOTAL QTY : 1 1 8.00 8.00 TOTAL QTY : TOTAL AMOUNT CASH CHANGE 8.00 10.00 2.00 G 0.45 THA... | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"manifest_index": 8,
"dataset_offset": 8,
"doc_type": "receipt",
"note": "Extended from the four SROIE Task 3 targets (company, date, address, total) to every extractable value on the page: company registration number, GST ID, time, cashier id, receipt... | null | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"index": 8,
"image_size": [
584,
1024
],
"normalisation": "pixel boxes divided by this shipped image's width/height",
"granularity": "line",
"verified": "36/36 sampled box texts found in the text GT",
"words": [
{
"text": "SWC ENTERPRISE SDN BHD"... | |
sroie_004 | easy | receipt | 3180301 SECURE PARKING CORPORATION S/B RIVERWALK VILLAGE BATU 3 51200 KL. GSTNO.000296652800 TEL NO: 1300881698 RECEIPT C13 RECEIPT NUMBER: K0131800235697 ENTRY TIME : 23.03.18 14:59 EXIT TIME: 23.03.18 15:27 PARK-DUR.: D:HH:MM 0:00:28 KIND OF PAYMENT: CASH PARKING FEE RM 0.94 ADD GST RM 0.06 *AMT RM 1.00 *RATE INCL. 6... | null | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"manifest_index": 11,
"dataset_offset": 11,
"doc_type": "receipt",
"note": "Extended from the four SROIE Task 3 targets (company, date, address, total) to every extractable value on this car-park exit ticket. The four-field schema captures almost nothi... | null | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"index": 11,
"image_size": [
447,
1127
],
"normalisation": "pixel boxes divided by this shipped image's width/height",
"granularity": "line",
"verified": "34/34 sampled box texts found in the text GT",
"words": [
{
"text": "3180301",
"bbox": [... | |
sroie_005 | easy | receipt | UROKO JAPANESE CUISINE SDN BHD 22A-1 SECTION 17 46400 PETALING JAYA SELANGOR. 03-7932 1023/0191 GST REG NO: 001126838272 BILL NO : 01H-25029 DATE : 20/03/2018 6:41:02 PM CASHIER : RESAN TABLE NO : S09 QTY DESCRIPTION AMOUNT 1 KO NABE INANIWA 24.00 SR UDON @ 24.00 1 GREEN TEA @ 1.00 1.00 SR 1 GYOZA 5PCS @ 15.00 15.00 SR... | {
"plain_text": "UROKO JAPANESE CUISINE SDN BHD 22A-1 SECTION 17 46400 PETALING JAYA SELANGOR. 03-7932 1023/0191 GST REG NO: 001126838272 BILL NO : 01H-25029 DATE : 20/03/2018 6:41:02 PM CASHIER : RESAN TABLE NO : S09 QTY DESCRIPTION AMOUNT 1 KO NABE INANIWA 24.00 SR UDON @ 24.00 1 GREEN TEA @ 1.00 1.00 SR 1 GYOZA 5P... | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"manifest_index": 14,
"dataset_offset": 14,
"doc_type": "receipt",
"note": "Extended from the four SROIE Task 3 targets (company, date, address, total) to every extractable value on this restaurant bill: phone, GST registration, bill number, full times... | null | {
"dataset": "jsdnrs/ICDAR2019-SROIE",
"split": "test",
"index": 14,
"image_size": [
627,
1071
],
"normalisation": "pixel boxes divided by this shipped image's width/height",
"granularity": "line",
"verified": "36/40 sampled box texts found in the text GT",
"words": [
{
"text": "UROKO JAPANESE CUISINE... | |
chart_000 | hard | chart | # Global Metal Prices (USD/ton, 2015–2020)
| Year | Copper | Steel |
|------|--------|-------|
| 2015 | 2000 | 1800 |
| 2016 | 1800 | 1600 |
| 2017 | 2100 | 1900 |
| 2018 | 2400 | 2200 |
| 2019 | 2300 | 2000 |
| 2020 | 2500 | 2300 |
**Source:** World Bank, London Metal Exchange | {
"plain_text": "# Global Metal Prices (USD/ton, 2015–2020)\n\n| Year | Copper | Steel |\n|------|--------|-------|\n| 2015 | 2000 | 1800 |\n| 2016 | 1800 | 1600 |\n| 2017 | 2100 | 1900 |\n| 2018 | 2400 | 2200 |\n| 2019 | 2300 | 2000 |\n| 2020 | 2500 | 2300 |\n\n**Source:** World Bank, London Metal ... | {
"dataset": "getomni-ai/ocr-benchmark",
"split": "test",
"manifest_index": 0,
"dataset_offset": 235,
"doc_type": "chart",
"json_schema": {
"type": "object",
"properties": {
"title": {
"type": "string",
"description": ""
},
"prices": {
"type": "array",
... | null | null | |
chart_001 | hard | chart | ## 2022 Education Industry Investment Analysis (Millions)
| | January | February | March |
| ----------------- | ------- | -------- | ----- |
| Preschool Education | 45.1 | 32.7 | 9.8 |
| K-12 Education | 52.6 | 38.4 | 14.2 |
| Higher Education | 57.8 | 41.5 ... | {
"plain_text": "## 2022 Education Industry Investment Analysis (Millions)\n\n| | January | February | March |\n| ----------------- | ------- | -------- | ----- |\n| Preschool Education | 45.1 | 32.7 | 9.8 |\n| K-12 Education | 52.6 | 38.4 | 14.2 |\n| Higher Educa... | {
"dataset": "getomni-ai/ocr-benchmark",
"split": "test",
"manifest_index": 1,
"dataset_offset": 357,
"doc_type": "chart",
"json_schema": {
"type": "object",
"properties": {
"title": {
"type": "string",
"description": ""
},
"eduction_investments": {
"type": ... | null | null | |
chart_002 | hard | chart | # Company structure
BD is structured to serve customers by providing unique solutions. The data below represents the company structure for FY 2019.
## Revenue by geography
(millions of dollars)
United States (including Puerto Rico)
$9,730
Europe
$3,359
Greater Asia (including Japan and Asia Pacific)
$2,726
Other... | {
"plain_text": "# Company structure\n\nBD is structured to serve customers by providing unique solutions. The data below represents the company structure for FY 2019.\n\n## Revenue by geography\n\n(millions of dollars)\n\nUnited States (including Puerto Rico)\n$9,730\n\nEurope\n$3,359\n\nGreater Asia (including Japa... | {
"dataset": "getomni-ai/ocr-benchmark",
"split": "test",
"manifest_index": 2,
"dataset_offset": 1,
"doc_type": "chart",
"json_schema": {
"type": "object",
"title": "Revenue by Geography and Segment",
"required": [
"geographyRevenue",
"productRevenue"
],
"properties": {
... | null | null | |
chart_003 | hard | chart | Figure 20: Global Workforce by Employment Level
| | 2020 | 2019 | 2018 | 2017 |
| :---------------------------- | :---- | :---- | :---- | :---- |
| Executive & Senior Management | 400 | 400 | 400 | 400 |
| Management | 1,000 | 1,000 | 1,000 | 1,000 |
| Profe... | {
"plain_text": "Figure 20: Global Workforce by Employment Level\n\n| | 2020 | 2019 | 2018 | 2017 |\n| :---------------------------- | :---- | :---- | :---- | :---- |\n| Executive & Senior Management | 400 | 400 | 400 | 400 |\n| Management | 1,000 | 1,000 |... | {
"dataset": "getomni-ai/ocr-benchmark",
"split": "test",
"manifest_index": 3,
"dataset_offset": 112,
"doc_type": "chart",
"note": "The getomni schema asks for one string (the largest employment category) and a four-row male/female series rounded to the nearest thousand — thirteen leaf values, and the round... | null | null | |
chart_004 | hard | chart | [uptime intelligence logo]
KEYNOTE REPORT
Uptime Institute Global Data Center Survey 2024
Figure 1
Cost issues are the top concern for management in 2024
Looking at the next 12 months, how concerned is your digital infrastructure management regarding each of the following issues? (n=638)
<table>
<tr>
<th></th>
... | {
"plain_text": "[uptime intelligence logo]\nKEYNOTE REPORT\nUptime Institute Global Data Center Survey 2024\n\nFigure 1\nCost issues are the top concern for management in 2024\nLooking at the next 12 months, how concerned is your digital infrastructure management regarding each of the following issues? (n=638)\n\n<t... | {
"dataset": "getomni-ai/ocr-benchmark",
"split": "test",
"manifest_index": 4,
"dataset_offset": 13,
"doc_type": "chart",
"json_schema": {
"type": "object",
"title": "DataCenterConcerns",
"required": [
"concerns"
],
"properties": {
"concerns": {
"type": "array",
... | null | null |
OCR Benchmark — Documents
The 93 document images and ground truth used by the ocr-benchmark harness. The benchmark code, the reference run results, and the full methodology live in the GitHub repo — this dataset is the document corpus only.
Structure
One train split, 93 rows, one row per document:
| Column | Type | Description |
|---|---|---|
image |
Image | The document page (PNG/JPG) |
stem |
string | Filename stem (e.g. invoice_000) |
tier |
string | Difficulty: easy, medium, or hard |
doc_type |
string | Document type (e.g. invoice, receipt, tax_form) |
gt_text |
string | Ground-truth plain text (null for 11 silver-GT docs) |
gt_json |
string | Serialized JSON — plain_text and table_html fields |
structured_gt |
string | Serialized JSON — json_schema and true_json (schema varies per doc type) |
markdown_gt |
string | Ground-truth markdown (null if unavailable) |
geometry_gt |
string | Bounding-box ground truth as serialized JSON (null if unavailable) |
JSON columns are stored as strings because their internal structure varies per
document type (a tax form's schema is not a receipt's schema). Parse with
json.loads().
Document types
| Tier | Types (count each) | Total |
|---|---|---|
| easy | receipt (6), invoice (6) | 12 |
| medium | bank_statement, shipping, tax_form, medical, payslip, lease, photo_receipt (5 each) | 35 |
| hard | form (6), handwritten (6), scanned_legacy (6), financial_table (5), chart (5), newspaper (6), legal (6), report (6) | 46 |
Where the documents come from
Every image is an excerpt from a public dataset. Documents were selected deterministically (fixed row indices), not cherry-picked by content.
| Source | Docs | License | GT provenance |
|---|---|---|---|
| ICDAR2019-SROIE (Huang et al.) | 6 | CC-BY-4.0 | Word annotations (human) |
| invoices-and-receipts_ocr_v1 | 6 | None declared | parsed_data field values (human-labeled) |
| OmniAI OCR Benchmark | 40 | MIT | true_markdown_output (human-annotated) |
| FUNSD (Jaume et al. 2019) | 6 | Non-commercial research/educational use | Word annotations (human) |
| IAM Handwriting (FKI/Univ. Bern) | 6 | Registration-required, non-commercial research | Line transcriptions (human); 15 lines collaged per page |
| IDL-WDS (UCSF Industry Documents) | 6 | Custom "idl-train" license | Legacy OCR annotations (machine-generated) |
| SynFinTabs | 5 | MIT | Exact synthetic cell text |
| RVL-CDIP (Harley et al.) | 18 | Research use | Silver — Tesseract at build time; judge-only evaluation |
FUNSD, IAM, and RVL-CDIP (30 of 93 docs) carry research-use or non-commercial restrictions. This dataset redistributes small excerpts solely for reproducible research benchmarking, with full attribution. If you are a rights holder and want a document removed, open an issue — it will be removed promptly.
Ground-truth provenance
- Human (75 docs) — source dataset annotations
- Model-verified (8 docs) — model-transcribed, cross-checked against 14-engine consensus
- Silver (11 docs) — machine-generated (Tesseract); scored by LLM judge only, never by text metrics
Per-document provenance, known limitations, and the full audit trail are in the
GitHub repo under
documents/ — and every benchmark run prints them before any score.
Usage
Load directly:
from datasets import load_dataset
import json
ds = load_dataset("ilsilfverskiold/ocr-benchmark", split="train")
row = ds[0]
row["image"] # PIL image
gt = json.loads(row["structured_gt"]) # json_schema + true_json
Or use it through the benchmark harness, which rebuilds the on-disk corpus layout and can rerun any engine against these documents:
git clone https://github.com/ilsilfverskiold/ocr-benchmark
cd ocr-benchmark
pip install -r requirements.txt
python scripts/download_data.py # fetches this dataset
python run_benchmark.py --engines tesseract,docling --tiers easy --skip-judge
The reference benchmark results (14 engines × 93 documents × 7 legs) ship with the GitHub repo — you only need this dataset if you want to rerun engines.
License
The images are excerpts from third-party datasets and keep their original licenses — see the source table above. The permissive subset (MIT / CC-BY-4.0) covers 57 of the 93 documents.
- Downloads last month
- 28