Deep Learning Models
In sports analytics, it's often important to catch the events. In volleyball, just like any other sport, there are a lot of gaps between the events. We need to catch the volleyball events when the game is in play, so to do this, we need to formulate it.
In this repo, we formulated it this way:
we sliced each video into 1-second clips and extracted each clip into its frames, which would become 30 frames each. Then we input them into a fine-tuned version of VideoMAE model on a custom dataset, and it outputs 1 label denoting service, in-play and no-play.
The label service is the state of the game that the player tosses the ball, and hits it to start the rally.
The label in-play denotes the game states that the players are playing, the event is in progress, and it hasn't been stopped.
The label no-play denotes the game states that there is no event or the players are just warming up and not playing for the score.
Dataset:
train:
| no-play | play | service |
|---|---|---|
| 7136 | 4886 | 2447 |
test:
| no-play | play | service |
|---|---|---|
| 794 | 544 | 273 |
Confusion matrix:
The evaluation results:
| eval_accuracy | eval_precision | eval_recall | eval_f1 | eval_loss |
|---|---|---|---|---|
| 0.9906890130353817 | 0.9911190984494304 | 0.9883168656922177 | 0.9896984571368996 | 2.413252830505371 |
Yolov8 is the state-of-the-art architecture in the object detection realm and it's developed by ultralytics group and is used for all object detection/segmentation tasks in this repo.
Datasets:
Dataset1:
| train set | test set |
|---|---|
| 23052 images | 857 images |
Dataset2:
The 2nd dataset is customizedly annotated by myself and it's a segmentation-type annotation.
| train set | test set |
|---|---|
| 3062 images | 384 images |
Datasets:
| train set | validation set |
|---|---|
| 4032 images | 448 images |
Dataset:
| train set | validation set | test set |
|---|---|---|
| 1952 images | 557 images | 279 images |
we use pre-trained yolov8n for the detection and tracking of players.