Datasets:
File size: 4,051 Bytes
67137fb eb3313e 14cc141 aeff35c 67137fb 14cc141 d3caa18 67137fb d3caa18 14cc141 67137fb 14cc141 d3caa18 14cc141 b38de6f 14cc141 d3caa18 14cc141 | 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 | ---
configs:
- config_name: all
data_files:
- split: train
path: "*-train.tar"
default: true
language: ins
license: cc-by-sa-4.0
datasets:
- bridgeconn/sign-dictionary-isl
tags:
- video
- parallel-corpus
- low-resource-languages
---
# Dataset Card for Sign Dictionary Dataset
This dataset contains Indian sign language videos with one gloss per video. There are 3077 seperate lex items or glosses included.
The dataset is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0).
## Dataset Details
There is a total of 2.5 hours of sign videos.
## Dataset Description
- Segmented sign videos
- Pose estimation data in the following formats
- skeletal video
- Frames wise body landmarks detected by dwpose as a numpy array
- Frames wise body landmarks detected by mediapose as .pose format
## How to use
```python
import webdataset as wds
import numpy as np
import json
import tempfile
import os
import cv2
def main():
buffer_size = 1024
dataset = (
wds.WebDataset(
"https://huggingface.co/datasets/bridgeconn/sign-dictionary-isl/resolve/main/shard_{00001..00002}-train.tar",
shardshuffle=False)
.shuffle(buffer_size)
.decode()
)
for sample in dataset:
''' Each sample contains:
'mp4',
'pose-dwpose.npz', 'pose-mediapipe.pose'
and 'json'
'''
# print(sample.keys())
# JSON metadata
json_data = sample['json']
print(json_data['filename'])
print(json_data['transcripts'])
print(json_data['glosses'])
# main video
mp4_data = sample['mp4']
process_video(mp4_data)
# dwpose results
dwpose_coords = sample["pose-dwpose.npz"]
frame_poses = dwpose_coords['frames'].tolist()
print(f"Frames in dwpose coords: {len(frame_poses)} poses")
print(f"Pose coords shape: {len(frame_poses[0][0])}")
print(f"One point looks like [x,y]: {frame_poses[0][0][0]}")
# mediapipe results in .pose format
pose_format_data = sample["pose-mediapipe.pose"]
process_poseformat(pose_format_data)
break
def process_poseformat(pose_format_data):
from pose_format import Pose
temp_file = None
try:
with tempfile.NamedTemporaryFile(suffix=".pose", delete=False) as tmp:
tmp.write(pose_format_data)
temp_file = tmp.name
data_buffer = open(temp_file, "rb").read()
pose = Pose.read(data_buffer)
print(f"Mediapipe results from pose-format: {pose.body.data.shape}")
except Exception as e:
print(f"Error processing pose-format: {e}")
finally:
if temp_file and os.path.exists(temp_file):
os.remove(temp_file) # Clean up the temporary file
def process_video(mp4_data):
print(f"Video bytes length: {len(mp4_data)} bytes")
temp_file = None
try:
# Processing video from temporary file
with tempfile.NamedTemporaryFile(suffix=".mp4", delete=False) as tmp:
tmp.write(mp4_data)
temp_file = tmp.name
cap = cv2.VideoCapture(temp_file)
if not cap.isOpened():
raise IOError(f"Could not open video file: {temp_file}")
# Example: Get video metadata
frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
fps = cap.get(cv2.CAP_PROP_FPS)
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
print(f"Video Info: {frame_count} frames, {fps:.2f} FPS, {width}x{height}")
# Example: Read and display the first frame (or process as needed)
ret, frame = cap.read()
if ret:
print(f"First frame shape: {frame.shape}, dtype: {frame.dtype}")
# You can then use this frame for further processing, e.g.,
frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
import matplotlib.pyplot as plt
plt.imshow(frame_rgb)
plt.title(f"Sample First Frame")
plt.show()
else:
print("Could not read first frame.")
cap.release()
except Exception as e:
print(f"Error processing external MP4: {e}")
finally:
if temp_file and os.path.exists(temp_file):
os.remove(temp_file) # Clean up the temporary file
if __name__ == '__main__':
main()
```
---
license: cc-by-sa-4.0
---
|