IPL - Datasets

IPL Azadi Soccer Dataset

Azadi Soccer Dataset is prepared for academic purposes by Image Processing Laboratory (IPL) members of Sharif University of Technology, headed by Dr. Mehran Fotouhi under the supervision of Prof. Shohreh Kasaei, in 12 May 2012. The dataset includes the video sequences that are captured from a domestic soccer matches in Iran Pro League. The prepared dataset is one of the few existing multiview soccer datasets which contains 9 fixed cameras and 3 broadcast cameras. The quality of videos is SD, the size of each frame is 720×576 and the frame rate is 25fps. The playfield size and the location of each camera are measured using a set of station theodolites. The dataset contains the actual video capturing phenomena (such as camera lens distortion, motion blur, lighting changes, shadows, small-size multiple moving objects, noise, and so forth). These make it almost ideal for being used as a practical dataset for research purposes. As such, several doctoral and master's theses have been defined at IPL. Some sample videos captured by the broadcast and fixed cameras are shown below. Several sub-datasets are also prepared for different research topics including ball detection, frame synchronization, player detection, and billboard ads. We also would like to appreciate the efforts of IPL members Mr. Mohamad Mehabadi and Mr. Arsh Aghababazadeh for their help on compressing and uploading the dataset. For more information please contact kasaei[at]sharif[dot]edu.

The soccer field model and the location of the moving camera (M₆), the 6 PTZ cameras (in both halftimes), and the 9 fixed cameras in the 1^st and 2^nd halftimes are shown below. For more info on camera setups please click here.

F^y_x(n): Fixed camera number x, in the y^th halftime, from the n^th part of the related source video.

P^y_x(n): PTZ camera number x, in the y^th halftime, from the n^th part of the related source video.

M^y_x(n): Moving camera number x, in the y^th halftime, from the n^th part of the related source video.

Soccer field model.

PTZ cameras in
full-match.

Fixed cameras in
1^st halftime.

Fixed cameras in
2^nd halftime.

F²₄

P²_{2,
center, longview}

F²₂

P²_{1,
offside}

F²₆

For citation purposes, you can use our publications.

Fixed Cameras

In IPL Azadi Soccer Dataset, the locations of fixed cameras have been changed at each halftime. In the first halftime, the whole playfield is covered by fixed cameras and in the second halftime, a half of the playfield is covered by all fixed cameras. Therefore, in the second configuration of fixed cameras, the number of views for a specific region of the playfield is more than that in the first configuration.

Camera Type	Camera Info	Total No. of Frames	Pose Changed from Initial Calibration	Synced Frame No.
F_2_1	F²₁(1)	45640	Yes	28671
F_2_2	F²₂	66750	No	34726
F_2_3	F²₃(2)	52001	No	19924
F_2_4	F²₄	72500	Yes	40350
F_2_5	F²₅(8)	43660	Yes	21810
F_2_6	F²₆	72501	Yes	40607
F_2_7	F²₇	72251	Yes	38231
F_2_8	F²₈	70251	No	40287
F_2_9	F²₉	72376	No	481397

To find the location of fixed cameras in each half of soccer game and other information please contact kasaei[at]sharif[dot]edu.

Ball Detection

This sub-dataset contains 2237 frames of 9 fixed cameras (totally 20133 frames), for which the coordination of ball at each view is determined manually and saved in a mat file. To reach this dataset and more information please contact kasaei[at]sharif[dot]edu.

F²₁

F²₂

F²₃

F²₄

F²₅

F²₆

F²₇

F²₈

F²₉

Frame Synchronization

This sub-dataset is about synchronization of multiple cameras in multiview camera applications. To reach this dataset and more information please contact kasaei[at]sharif[dot]edu.

F²₃

F²₄

F²₄ (Dropped Frame)

F²₄

Broadcast Cameras

The locations of broadcast cameras in Azadi stadium are standard. In the Azadi dataset, the videos of 2 offside cameras and 1 center camera are available as non-directed broadcast cameras. The camera parameters are varying during the match.

Synced Camera Info	Start Time
P²_{1, offside}(3)	0h, 26m, 57s
P²_{2, center, longview}	1h, 11m, 11s
M²_{6, crane}(4)	1h, 10m, 45s

To find the location of broadcast cameras in Azadi soccer dataset and other information please contact kasaei[at]sharif[dot]edu.

Player Detection

The player detection dataset contains 6 different sequences of center camera. Each sequence has different number of frames which are gathered from different views. To reach this dataset and more information please contact kasaei[at]sharif[dot]edu.

P²_{2,
center, longview}

P¹_{2,
center, longview}

P²_{2,
center, longview}

Billboard Ads

This sub-dataset, the coordinates of the 4 corners of a advertisement billboards are determined manually for a sequence of center camera. To reach this dataset and more information please contact kasaei[at]sharif[dot]edu.

Related Publications

M. Fotouhi, S. Fouladi, and S. Kasaei. Projection Matrix by Orthogonal Vanishing Points. Springer, Multimedia Tools & Applications, 76(15):16189-16223, August 2017. [BibTeX]
@article{fotouhi2017projection, title={Projection Matrix by Orthogonal Vanishing Points}, author={Fotouhi, Mehran and Fouladi, Sadjad and Kasaei, Shohreh}, journal={Springer, Multimedia Tools and Applications}, volume = "76", number = "15", pages = "16189-16223", month = "August", year={2017}, }
A. Bozorgpour, M. Fotouhi, and S. Kasaei. Robust Homography Optimization in Soccer Scenes. In The 23rd Iranian Conference on Electrical Engineering (ICEE), pages 787-792, Tehran, Iran, May 2015. [BibTeX]
@inproceedings{bozorgpour2015robust, title={Robust homography optimization in soccer scenes}, author={Bozorgpour, Afshin and Fotouhi, Mehran and Kasaei, Shohreh}, booktitle={2015 23rd Iranian Conference on Electrical Engineering}, pages={787--792}, year={2015}, organization={IEEE} }
M. Fotouhi, A. Bozorgpour, and S. Kasaei. Automatic Soccer Field Line Recognition by Minimum Information. In International Symposium on Artificial Intelligence & Signal Processing (AISP), pages 136-142, Mashhad, Iran, March 2015. [BibTeX]
@inproceedings{fotouhi2015automatic, title={Automatic soccer field line recognition by minimum information}, author={Fotouhi, Mehran and Bozorgpour, Afshin and Kasaei, Shohreh}, booktitle={Artificial Intelligence and Signal Processing (AISP), 2015 International Symposium on}, pages={136--142}, year={2015}, organization={IEEE} }