Match Video Tensors

If you use these data, please remember to cite the following paper:<br>Sorano, D., Carrara, F., Cintia, P., Falchi, F., Pappalardo, L. (2020) Automatic Pass Annotation from Soccer VideoStreams Based on Object Detection and LSTM, In: Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2020.<br><br>Tensors extracted from soccer video broadcasts. Each file is a zip of folder and corresponds to a single half of a match. Each file in the folder (in .pickle format) corresponds to a frame of the video.<br>This item contains the following files/matches:- <b>roma_juve_1H_tensors.zip</b>: tensors/frames of the first half of match Roma vs Juventus- <b>roma_juve_2H_tensors.zip</b>: tensors/frames of the second half of match Roma vs Juventus- <b>roma_lazio_1H_tensors.zip</b>: tensors/frames of the first half of match Roma vs Lazio- <b>sassuolo_inter_1H_tensors.zip</b>: tensors/frames of the first half of match Sassuolo vs Inter- <b>sassuolo_inter_2H_tensors.zip</b>: tensors/frames of the second half of match Sassuolo vs Inter<br><br>