Pose-Projected Action Recognition Hourglass Network (PARHN) in Soccer
Bibliographic record
Abstract
Current research on soccer action recognition does not focus on player-level actions. We introduce the pose-projected action recognition hourglass network (PARHN) for performing player-level action recognition in soccer. This network is inspired by ARHN, a network originally introduced for hockey action recognition. PARHN has two main novelties in its structure. First, it includes an embedded pose projection component that regularizes the numerical range of the player's pose vector, by applying two separate zero-phase component analysis (ZCA)-whitening. The projected pose vector is ef-fectively learned by few succeeding layers, and significantly improves the performance of the network and its generalizationability. Second, PARHN incorporates the temporal information by having a parallel structure for extracting projected pose vectors from all frames of an input sequence and also by using Long short-term memory (LSTM) layers to integrate the pose vectors across the input frames. Also, a new dataset, named SAR4 (standing for, Soccer Action Recognition for 4 action types), is generated. It includes 1292 video sequences, in which soccer players are tracked and labeled for performing four types of action (i.e., goalkeeper diving, player shooting, receiving pass and giving pass). Introduced network achieves the overall F1-score of 88.1% on the test data, which is 20.1% better than the result of the baseline network, ARHN.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".