Robust Joint Audio–Video Talker Localization in Video Conferencing Using Reliability Information—II: Bayesian Network Fusion
Bibliographic record
Abstract
This paper builds on our August 2004 paper where data fusion is used to combine results from multiple audio and video localizers in order to locate a talker in video conferencing applications. The two studies differ in the type of data fusion engine used. The former study explored using a summing voter for data fusion, whereas this study employs a Bayesian network. The novelty of both papers is the use of reliability estimates to improve the overall localization performance and robustness. Reliability estimates, which are derived based on known physical properties of each individual localizer, were used to gauge the trustworthiness of the current localization results. These reliability estimates were introduced into the fusion engines to achieve better performance. Although the summing voter fusion engine used in improves the overall localization performance, it does not take into account the unique characteristics of each localizer. The Bayesian network allows the inclusion of these unique characteristics as part of the fusion process and therefore further improves the overall localization performance. When one or more localizers fails, the persistent erroneous data streams from the failed localizers can negatively affect a statistically based data fusion method, like the Bayesian network, resulting in poor localization accuracy . The study described in this paper explores the use of the reliability estimates to automatically stop the failed devices from contributing in the data fusion process, hence improving the overall robustness of the system. In this paper, we investigate the impact of 1) using a Bayesian network as the data fusion engine, 2) adding reliability estimates into the fusion engine, and 3) using reliability estimates to stop the failed localizers from contributing in the fusion process.
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 | 0.006 | 0.018 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".