From d514784878b16ee6af0824da1bfb067b3ed7838a Mon Sep 17 00:00:00 2001
From: Xinyu Wang <45810070+xinke-wang@users.noreply.github.com>
Date: Tue, 15 Nov 2022 19:42:43 +0800
Subject: [PATCH] [Feature] Add SVTP to dataset preparer (#1523)

* add svtp

* fix comments
---
 dataset_zoo/svtp/metafile.yml   | 26 ++++++++++++++++++++++++++
 dataset_zoo/svtp/sample_anno.md |  7 +++++++
 dataset_zoo/svtp/textrecog.py   | 27 +++++++++++++++++++++++++++
 3 files changed, 60 insertions(+)
 create mode 100644 dataset_zoo/svtp/metafile.yml
 create mode 100644 dataset_zoo/svtp/sample_anno.md
 create mode 100644 dataset_zoo/svtp/textrecog.py

diff --git a/dataset_zoo/svtp/metafile.yml b/dataset_zoo/svtp/metafile.yml
new file mode 100644
index 00000000..a544d35d
--- /dev/null
+++ b/dataset_zoo/svtp/metafile.yml
@@ -0,0 +1,26 @@
+Name: 'Street View Text Perspective (SVT-P)'
+Paper:
+  Title: Recognizing Text with Perspective Distortion in Natural Scenes
+  URL: https://openaccess.thecvf.com/content_iccv_2013/papers/Phan_Recognizing_Text_with_2013_ICCV_paper.pdf
+  Venue: ICCV
+  Year: '2013'
+  BibTeX: '@inproceedings{phan2013recognizing,
+  title={Recognizing text with perspective distortion in natural scenes},
+  author={Phan, Trung Quy and Shivakumara, Palaiahnakote and Tian, Shangxuan and Tan, Chew Lim},
+  booktitle={Proceedings of the IEEE International Conference on Computer Vision},
+  pages={569--576},
+  year={2013}}'
+Data:
+  Website: http://www.comp.nus.edu.sg/~phanquyt/
+  Language:
+    - English
+  Scene:
+    - Natural Scene
+  Granularity:
+    - Word
+  Tasks:
+    - textrecog
+  License:
+    Type: N/A
+    Link: N/A
+  Format: .txt
diff --git a/dataset_zoo/svtp/sample_anno.md b/dataset_zoo/svtp/sample_anno.md
new file mode 100644
index 00000000..11711e25
--- /dev/null
+++ b/dataset_zoo/svtp/sample_anno.md
@@ -0,0 +1,7 @@
+**Text Recognition**
+
+```txt
+13_15_0_par.jpg WYNDHAM
+13_15_1_par.jpg HOTEL
+12_16_0_par.jpg UNITED
+```
diff --git a/dataset_zoo/svtp/textrecog.py b/dataset_zoo/svtp/textrecog.py
new file mode 100644
index 00000000..ca0cc0ae
--- /dev/null
+++ b/dataset_zoo/svtp/textrecog.py
@@ -0,0 +1,27 @@
+data_root = 'data/svtp'
+cache_path = 'data/cache'
+
+data_obtainer = dict(
+    type='NaiveDataObtainer',
+    cache_path=cache_path,
+    data_root=data_root,
+    files=[
+        dict(
+            url='https://download.openmmlab.com/mmocr/data/svtp.zip',
+            save_name='svtp.zip',
+            md5='4232b46c81ba99eea6d057dcb06b8f75',
+            split=['test'],
+            content=['image', 'annotation'],
+            mapping=[['svtp/par1', 'textrecog_imgs/test'],
+                     ['svtp/gt.txt', 'annotations/test.txt']]),
+    ])
+
+data_converter = dict(
+    type='TextRecogDataConverter',
+    splits=['test'],
+    data_root=data_root,
+    gatherer=dict(type='mono_gather', mapping="f'{split}.txt'"),
+    parser=dict(
+        type='ICDARTxtTextRecogAnnParser', separator=' ', format='img text'),
+    dumper=dict(type='JsonDumper'),
+    delete=['svtp', 'annotations'])