CrowdTranscoding: Online Video Transcoding With Massive Viewers
Bibliographic record
Abstract
Driven by the advances in personal computing devices and the prevalence of high-speed network accesses, crowdsourced livecast platforms have emerged in recent years, through which numerous broadcasters lively stream their video content to fellow viewers. Compared to professional video producers and broadcasters, these new generation broadcasters are highly heterogeneous in terms of the network/system configurations and, therefore, the generated video quality, which calls for massive encoding and transcoding in order to unify the video sources and serve multiple quality versions to viewers with different configurations. On the other hand, with the rapid evolution in the hardware industry, high-performance processors become mainstream in personal computer market. More end devices can easily transcode high-quality videos in realtime. We witness huge computational resource among the massive fellow viewers that could potentially be used for transcoding. In this paper, we propose CrowdTranscoding, a novel framework for crowdsourced livecast systems that offloads the transcoding assignment to the massive viewers. We identify that the key challenges in CrowdTranscoding are to detect qualified stable viewers and to properly assign them to the source channels. We put forward a viewer crowdsourcing transcode scheduler to smartly schedule the workload assignment. Our solution has been evaluated under diverse viewer/channel conditions as well as different parameter settings. The trace-driven simulation confirms the superiority of CrowdTranscoder, while our PlanetLab-based and real world end-viewer experiments show the practical performance of our approach, which also give hint to the further enhancement.
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 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".