Automatic audience-informed video summarization of hockey broadcasts
Bibliographic record
Abstract
In the past decade, online video has risen to become one of the most common ways of sharing and consuming media.With so much video content readily-available, browsing or processing these data is difficult due to the long duration of a typical video.Video summaries, on the other hand, provide an effective way for viewers to enjoy the highlights of a video without having to watch the entire video.In this work, we propose a system that automatically summarizes hockey broadcasts into highlight videos.We leverage the audience of a game-both the broadcasters and the crowd-to decide which shots should be included in the final highlight reel.We argue that since the game is directly sourced by the crowd and the broadcasters, our direct approach is more effective than similar techniques in the literature which rely on external annotators.Our system accomplishes this by first extracting video and audio features using convolutional neural networks (CNN) and searching for key indicators of excitement (e.g.cheering or celebration).Lastly, the best highlights are dynamically selected under a set of constraints (such as output duration) and composited into the final summary.Our results demonstrate that this technique accurately extracts the most exciting moments of a hockey game while avoiding the bias from using predetermined metrics or external annotations.i Abrégé Dans le passé, les vidéos en ligne ont pris de l'ampleur pour devenir une des façons les plus communes pour partager et consommer les médias.Avec autant de contenu vidéo disponible, la consultation ou le traitement de ces données est difficile vu la longueur d'un vidéo typique.Les résumés vidéos, d'un autre côté, fournissent une façon efficace pour les consommateurs d'apprécier les points importants d'un vidéo sans devoir le regarder au complet.Dans ce travail, nous proposons un système qui résume automatiquement les émissions de hockey en un court vidéo des points forts.Nous tirons avantage du public d'un match, à la fois des spectateurs de l'émission et la foule, pour décider quelles parties devraient être incluses dans la bande des points forts.Nous affirmons que puisque la partie s'inspire directement de la foule et des spectateurs, notre approche directe est plus efficace que les techniques similaires dans la littérature qui se fit à des commentateurs extérieurs.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
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; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".