Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
This contains 16 frames of the "eating-spaghetti" video of the Kinetics-400 dataset, with the following frame indices being used: [164 168 172 176 181 185 189 193 198 202 206 210 215 219 223 227].
|
| 2 |
+
|
| 3 |
+
This is the code:
|
| 4 |
+
|
| 5 |
+
```
|
| 6 |
+
from decord import VideoReader, cpu
|
| 7 |
+
from huggingface_hub import hf_hub_download
|
| 8 |
+
import numpy as np
|
| 9 |
+
|
| 10 |
+
file_path = hf_hub_download(
|
| 11 |
+
repo_id="nielsr/video-demo", filename="eating_spaghetti.mp4", repo_type="dataset"
|
| 12 |
+
)
|
| 13 |
+
vr = VideoReader(file_path, num_threads=1, ctx=cpu(0))
|
| 14 |
+
|
| 15 |
+
# get 16 frames
|
| 16 |
+
vr.seek(0)
|
| 17 |
+
indices = [164 168 172 176 181 185 189 193 198 202 206 210 215 219 223 227]
|
| 18 |
+
|
| 19 |
+
# create a list of NumPy arrays
|
| 20 |
+
video = [buffer[i] for i in range(buffer.shape[0])]
|
| 21 |
+
|
| 22 |
+
video_numpy = np.array(video)
|
| 23 |
+
with open('spaghetti_video.npy', 'wb') as f:
|
| 24 |
+
np.save(f, video_numpy)
|
| 25 |
+
```
|