File size: 6,822 Bytes
386b31f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 |
import os
from pathlib import Path, PureWindowsPath
from typing import Dict, List, Tuple
try:
import cv2
except:
print("Install the `cv2` package to use.")
import datasets
import pandas as pd
from seacrowd.utils import schemas
from seacrowd.utils.configs import SEACrowdConfig
from seacrowd.utils.constants import Licenses, Tasks
_CITATION = """\
@article{tupal4476867fsl105,
title={FSL105: The Video Filipino Sign Language Sign Database of Introductory 105 FSL Signs},
author={Tupal, Isaiah Jassen Lizaso and Melvin, Cabatuan K},
journal={Available at SSRN 4476867}
}
"""
_DATASETNAME = "fsl_105"
_DESCRIPTION = """\
FSL-105 is a video dataset for 105 different Filipino Sign Language (FSL) signs.
Each sign is categorized into one of 10 categories and is each represented by approximately 20 four-second video samples.
Signs were performed by adult deaf FSL signers on a blank blue background and reviewed by an FSL expert.
"""
_HOMEPAGE = "https://data.mendeley.com/datasets/48y2y99mb9/2"
_LICENSE = Licenses.CC_BY_4_0.value
_LOCAL = False
_URLS = {
"clips": "https://prod-dcd-datasets-public-files-eu-west-1.s3.eu-west-1.amazonaws.com/de95a3c3-02f4-4a3f-9a9e-ce2371160275",
"train": "https://prod-dcd-datasets-public-files-eu-west-1.s3.eu-west-1.amazonaws.com/09c71779-3a2a-4c98-8d9b-0ef74f54d92a",
"test": "https://prod-dcd-datasets-public-files-eu-west-1.s3.eu-west-1.amazonaws.com/39af8117-6b44-47b9-a551-0bdc40837295",
}
_LANGUAGES = ["psp"]
_SUPPORTED_TASKS = [Tasks.VIDEO_TO_TEXT_RETRIEVAL, Tasks.VIDEO_CAPTIONING]
_SOURCE_VERSION = "1.0.0"
_SEACROWD_VERSION = "2024.06.20"
class FSL105Dataset(datasets.GeneratorBasedBuilder):
"""
FSL-105 is a video dataset for 105 different Filipino Sign Language (FSL) signs.
Each sign is categorized into one of 10 categories and is each represented by approximately 20 four-second video samples.
Signs were performed by adult deaf FSL signers on a blank blue background and reviewed by an FSL expert.
"""
SOURCE_VERSION = datasets.Version(_SOURCE_VERSION)
SEACROWD_VERSION = datasets.Version(_SEACROWD_VERSION)
BUILDER_CONFIGS = [
SEACrowdConfig(
name=f"{_DATASETNAME}_source",
version=SOURCE_VERSION,
description=f"{_DATASETNAME} source schema",
schema="source",
subset_id=f"{_DATASETNAME}",
),
SEACrowdConfig(
name=f"{_DATASETNAME}_seacrowd_vidtext",
version=SEACROWD_VERSION,
description=f"{_DATASETNAME} SEACrowd schema",
schema="seacrowd_vidtext",
subset_id=f"{_DATASETNAME}",
),
]
DEFAULT_CONFIG_NAME = f"{_DATASETNAME}_source"
category = [
"CALENDAR",
"COLOR",
"DAYS",
"DRINK",
"FAMILY",
"FOOD",
"GREETING",
"NUMBER",
"RELATIONSHIPS",
"SURVIVAL",
]
def _info(self) -> datasets.DatasetInfo:
if self.config.schema == "source":
features = datasets.Features(
{
"id": datasets.Value("string"),
"video_path": datasets.Value("string"),
"text": datasets.Value("string"),
"labels": datasets.ClassLabel(names=self.category),
"metadata": {
"resolution": {
"width": datasets.Value("int64"),
"height": datasets.Value("int64"),
},
"duration": datasets.Value("float32"),
"fps": datasets.Value("float32"),
},
}
)
elif self.config.schema == "seacrowd_vidtext":
features = schemas.video_features
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=features,
homepage=_HOMEPAGE,
license=_LICENSE,
citation=_CITATION,
)
def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
"""Returns SplitGenerators."""
clips = dl_manager.download_and_extract(_URLS["clips"])
train = dl_manager.download_and_extract(_URLS["train"])
test = dl_manager.download_and_extract(_URLS["test"])
train_df = pd.read_csv(train)
test_df = pd.read_csv(test)
return [
datasets.SplitGenerator(
name=datasets.Split.TRAIN,
gen_kwargs={
"filepath": {
"clips": clips,
"data": train_df,
},
"split": "train",
},
),
datasets.SplitGenerator(
name=datasets.Split.TEST,
gen_kwargs={
"filepath": {"clips": clips, "data": test_df},
"split": "test",
},
),
]
def _generate_examples(self, filepath: Path, split: str) -> Tuple[int, Dict]:
"""Yields examples as (key, example) tuples."""
for key, example in filepath["data"].iterrows():
video = cv2.VideoCapture(os.path.join(filepath["clips"], PureWindowsPath(example["vid_path"]).as_posix()))
fps = video.get(cv2.CAP_PROP_FPS)
frame_count = video.get(cv2.CAP_PROP_FRAME_COUNT)
duration = frame_count / fps
vid_width = video.get(cv2.CAP_PROP_FRAME_WIDTH)
vid_height = video.get(cv2.CAP_PROP_FRAME_HEIGHT)
if self.config.schema == "source":
yield key, {
"id": str(key),
"video_path": os.path.join(filepath["clips"], example["vid_path"]),
"text": example["label"],
"labels": example["category"],
"metadata": {
"resolution": {
"width": vid_width,
"height": vid_height,
},
"duration": duration,
"fps": fps,
},
}
elif self.config.schema == "seacrowd_vidtext":
yield key, {
"id": str(key),
"video_path": os.path.join(filepath["clips"], example["vid_path"]),
"text": example["label"],
"metadata": {
"resolution": {
"width": vid_width,
"height": vid_height,
},
"duration": duration,
"fps": fps,
},
}
|