PadelVic

Padel-specific collection of datasets targeted to train and test computer vision applications on padel videos

Captured datasets

Dataset	Height	Device	Resolution	FPS	Format	Link
Main camera	6 m	Panasonic AG-UX180 4K	3840x2160	50	MP4	3626x1960 version, 5.2 GB
Second camera	7.5 m	GoPro Hero9 Black	3840x2160	60	MP4	2806x1870 version, 5.2 GB
Court level 1	1.45 m	Samsung Galaxy S22+	1920x1080	60	MP4	1920x1080 version, 3.2 GB
Court level 2	1.45 m	iPhone 13 Pro Max	1920x1080	60	MOV	1920x1080 version, 3.2 GB
Motion Capture	N/A	Xsens MVN Awinda	N/A	60	BVH	One player, all games, 183 MB

Synthethic videos (and CSV files with ground truth positions)

Dataset	Source	Format	Link
1 player @bottom, no court	Xsens MVN Awinda	MP4	Sample, 29 MB
1 player @bottom, court, GT position sphere	Xsens MVN Awinda	MKV	Sample, 9 MB
1 player @bottom, court	Xsens MVN Awinda	MKV	001: Two rallies, 88 MB CSV with GT positions
1 player @bottom, court	Xsens MVN Awinda	MKV	002: One rally, 61 MB CSV with GT positions
1 player @top, court	Xsens MVN Awinda	MKV	002: One rally, 60 MB CSV with GT positions

Derived data (human detection and pose estimation from video)

Dataset	Source	Format	Link
Player tracking results	Panasonic AG-UX180 4K	MP4	Sample, Game 2, 16MB
Player positions	Panasonic AG-UX180 4K	XLSX	Sample, 550 KB
Player poses	Panasonic AG-UX180 4K	JSON	Sample, 74 MB

Note

The player positions and player poses datasets above are just a sample, computed with a work-in-progress, non-robust player tracking method.

How to cite

"PADELVIC: Multicamera videos and motion capture data of an amateur padel match", by Mohammadreza Javadiha, Carlos Andujar, Michele Calvanese, Enrique Lacasa, Jordi Moyés, Jose Luis Ponton, Antonio Susín and Jiabo Wang. Padel Scientific Journal 2 (1), 89-106, 2024. DOI 10.17398/2952-2218.2.89

Licence

This dataset can be used only for research purposes. More details on the CC-BY-like license coming soon.