Visualisasi: Jelajahi Tahu Data Anda
Keterangan:
Kitti berisi serangkaian tugas penglihatan yang dibuat menggunakan platform mengemudi otonom. Benchmark lengkap berisi banyak tugas seperti stereo, aliran optik, odometri visual, dll. Dataset ini berisi dataset deteksi objek, termasuk gambar monokular dan kotak pembatas. Dataset berisi 7481 gambar pelatihan yang dianotasi dengan kotak pembatas 3D. Deskripsi lengkap anotasi dapat ditemukan di readme dari readme kit pengembangan objek di beranda Kitti.
Homepage: http://www.cvlibs.net/datasets/kitti/
Source code:
tfds.object_detection.Kitti
versi:
-
3.1.0
: Tidak ada catatan rilis. -
3.2.0
(default): devkit diperbarui.
-
Ukuran download:
11.71 GiB
Ukuran dataset:
5.27 GiB
Auto-cache ( dokumentasi ): Tidak ada
Splits:
Membelah | Contoh |
---|---|
'test' | 711 |
'train' | 6.347 |
'validation' | 423 |
- fitur:
FeaturesDict({
'image': Image(shape=(None, None, 3), dtype=tf.uint8),
'image/file_name': Text(shape=(), dtype=tf.string),
'objects': Sequence({
'alpha': tf.float32,
'bbox': BBoxFeature(shape=(4,), dtype=tf.float32),
'dimensions': Tensor(shape=(3,), dtype=tf.float32),
'location': Tensor(shape=(3,), dtype=tf.float32),
'occluded': ClassLabel(shape=(), dtype=tf.int64, num_classes=4),
'rotation_y': tf.float32,
'truncated': tf.float32,
'type': ClassLabel(shape=(), dtype=tf.int64, num_classes=8),
}),
})
Kunci diawasi (Lihat
as_supervised
doc ):None
Gambar ( tfds.show_examples ):
- Contoh ( tfds.as_dataframe ):
- Citation:
@inproceedings{Geiger2012CVPR,
author = {Andreas Geiger and Philip Lenz and Raquel Urtasun},
title = {Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite},
booktitle = {Conference on Computer Vision and Pattern Recognition (CVPR)},
year = {2012}
}