Hugging Face's logo

Datasets:
XSpaceCoderX
/
ACE-Rallies

Tasks:
Tabular Classification
Tabular Regression
Image-to-3D
Modalities:
Text
Size:
< 1K
ArXiv:
License:
Dataset card Data Studio Files
xet
 Community
Dataset Viewer
Auto-converted to Parquet Duplicate
game
string
clip
string
start
string
end
string
1
1
0
205
1
2
5
114
1
3
3
47
1
4
0
102
1
5
0
138
1
6
6
52
1
7
0
437
1
8
2
45
1
9
2
113
1
10
2
105
1
11
3
124
1
12
4
118
1
13
0
252
2
1
0
349
2
2
0
198
2
3
0
33
2
4
0
44
2
5
0
195
2
6
0
550
2
7
0
186
2
8
0
632
3
1
0
176
3
2
0
28
3
3
0
397
3
4
0
416
3
5
0
130
3
6
0
236
3
7
0
311
3
8
0
46
3
9
0
136
4
1
0
597
4
2
0
484
4
3
0
111
4
4
0
101
4
5
0
224
4
6
0
232
4
7
0
482
5
1
9
190
5
2
3
52
5
3
12
69
5
4
9
246
5
5
0
109
5
6
5
220
5
7
14
189
5
8
0
116
5
9
0
180
5
10
0
109
5
11
0
34
5
12
0
172
5
13
0
40
5
14
0
34
5
15
0
173
6
1
92
587
6
2
16
578
6
3
176
528
6
4
61
267
7
1
0
127
7
2
8
201
7
3
6
45
7
4
29
871
7
5
17
203
7
6
0
104
7
8
9
53
7
9
5
96
8
1
0
117
8
2
0
91
8
3
0
382
8
4
0
132
8
5
0
78
8
6
0
65
8
7
0
421
8
8
0
318
8
9
0
237
9
1
0
82
9
2
0
257
9
3
0
358
9
4
0
105
9
5
0
291
9
6
0
107
9
7
0
86
9
8
0
55
9
9
0
222
10
1
0
315
10
2
0
131
10
3
0
160
10
4
0
46
10
5
0
40
10
6
0
291
10
7
0
122
10
8
0
48
10
9
0
48
10
10
0
165
10
12
0
585

Paper in the making

ACE-Rallies Dataset

This dataset was created for the Master's thesis "From Broadcast to 3D: A Deep Learning Approach for Tennis Trajectory and Spin Estimation" by Alexandra Göppert at the University Augsburg, Chair of Machine Learning and Computer Vision.

This datasets serves as an enriched version of the original TrackNet Tennis dataset, supplementing the original broadcast tracking data with manual ball spin annotations and 2D human pose estimations for the hitting player. It was specifically designed to be used as a validation and test set (split 33% to 66% respectively) for the 2D-to-3D trajectory uplifting models available in the tennisuplifting GitHub repository. This dataset does not contain any images but uses primarily coordinates for indicating the position of the ball, the 16 court keypoints (see image) as well as the 17 COCO Whole Body keypoints (see end of file) of the hitting player on the respective image.

Court Keypoints

Original TrackNet Data

The foundation of this dataset is the TrackNet dataset, which consists of broadcast tennis videos and 2D ball tracking data. The original dataset was introduced in this paper and downloaded from Kaggle.

Folder Structure

The repository is organized into 93 sequentially numbered folders, ranging from rally_0000 to rally_0092. Each folder represents a single, continuous tennis rally (a sequence of consecutive shots starting with a serve or ball toss).

Data Structures per Rally

Inside each rally_xxxx folder, you will find exactly seven separate .npy files and one info.json file containing the isolated data for that specific sequence.

The JSON File

  • info.json: This file contains the essential metadata linking this rally back to the original TrackNet data. The fields include the specific game and clip the trajectory was cut from, as well as the exact start frame number and end frame number relative to the original TrackNet video.

The NPY Files

The seven numpy arrays store the spatial, temporal, and pose data for the rally:

  • Mint.npy: A [3 x 3] array representing the reverse-engineered intrinsic camera matrix.
  • Mext.npy: A [3 x 4] array representing the reverse-engineered extrinsic camera matrix (containing the rotation matrix and translation vector).
  • times.npy: A 1D array of length [T] (where T is the total number of frames in the rally) containing the exact timestamps for each frame.
  • r_img.npy: A [T x 3] array containing the 2D ball tracking data per frame. The three columns represent the u (horizontal) coordinate, the v (vertical) coordinate, and a visibility class. The visibility class is binary, where 0 means the ball is out of frame/occluded, 1 means visible or hard to spot. The visibility class was directly extracke out of the TrackNet Dataset.
  • 2dPoseEstimation.npy: A [17 x 3] array containing the 2D human pose estimation of the hitting player. The rows correspond to the 17 COCO-WholeBody keypoints, and the columns represent the u coordinate, v coordinate, and a model confidence score. For whole rallies, this pose is captured at the specific frame where the ball leaves the server's hand (or the first frame if the toss isn't visible).
  • spin_class_per_shot.npy & spin_class_per_frame.npy: These files map the initial ball spin of the shots. The classes are categorized as 1 (topspin), 2 (backspin), and 0 (no spin), with 0 typically used for the initial ball toss. spin_class_per_shot provides an array of length [S] mapping one spin class to each of the S shots in the rally. spin_class_per_frame.npy has the length [T], which assigns each frame of the video a spin class according to the initial spin value for the respective shot.
  • new_trajectory_frame_idx.npy: An array of length [S] providing the exact frame indices (time steps) at which each new trajectory begins and the corresponding new spin value can be measured.

Download and Usage

This dataset utilizes Git LFS (Large File Storage) for binary .npy files. To ensure that the actual data is downloaded instead of small text pointers, please use the huggingface_hub library.

Installation 

pip install huggingface_hub numpy

Dowload the full dataset 

from huggingface_hub import snapshot_download

snapshot_download(
    repo_id="XSpaceCoderX/ACE-Rallies", 
    repo_type="dataset", 
    local_dir="./data"
)

Loading the data 

import numpy as np

# Example: Loading a specific file after download
data = np.load("./data/path/to/file.npy", allow_pickle=True)
print(f"Data shape: {data.shape}")
COCO WholeBody keypoints
Downloads last month
-
Number of rows:
93
Total file size:
1.31 MB

Papers for XSpaceCoderX/ACE-Rallies