Papers › COMEDIAN: Self-Supervised Learning and Knowledge Distillation for Action Spotting...
COMEDIAN: Self-Supervised Learning and Knowledge Distillation for Action Spotting using Transformers
Julien Denize, Mykola Liashuha, Jaonary Rabarisoa, Astrid Orcesi, Romain Hérault
We present COMEDIAN, a novel pipeline to initialize spatiotemporal transformers for action spotting, which involves self-supervised learning and knowledge distillation. Action spotting is a timestamp-level temporal action detection task. Our pipeline consists of three steps, with two initialization stages. First, we perform self-supervised initialization of a spatial transformer using short videos as input. Additionally, we initialize a temporal transformer that enhances the spatial transformer's outputs with global context through knowledge distillation from a pre-computed feature bank aligned with each short video segment. In the final step, we fine-tune the transformers to the action spotting task. The experiments, conducted on the SoccerNet-v2 dataset, demonstrate state-of-the-art performance and validate the effectiveness of COMEDIAN's pretraining paradigm. Our results highlight several advantages of our pretraining pipeline, including improved performance and faster convergence compared to non-pretrained models.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Action Spotting | SoccerNet-v2 | COMEDIAN (ViSwin T ens.) | Average-mAP | 77.6 | #1 of 10 | Archive leaderboard | report |
| Action Spotting | SoccerNet-v2 | COMEDIAN (ViSwin T ens.) | Tight Average-mAP | 73.1 | #1 of 10 | Archive leaderboard | report |
| Action Spotting | SoccerNet-v2 | COMEDIAN (ViViT T ens.) | Average-mAP | 77.1 | #2 of 10 | Archive leaderboard | report |
| Action Spotting | SoccerNet-v2 | COMEDIAN (ViViT T ens.) | Tight Average-mAP | 72.0 | #2 of 10 | Archive leaderboard | report |
| Action Spotting | SoccerNet-v2 | COMEDIAN (ViSwin T) | Average-mAP | 76.6 | #3 of 10 | Archive leaderboard | report |
| Action Spotting | SoccerNet-v2 | COMEDIAN (ViSwin T) | Tight Average-mAP | 71.6 | #3 of 10 | Archive leaderboard | report |
| Action Spotting | SoccerNet-v2 | COMEDIAN (ViViT T) | Average-mAP | 76.1 | #4 of 10 | Archive leaderboard | report |
| Action Spotting | SoccerNet-v2 | COMEDIAN (ViViT T) | Tight Average-mAP | 70.7 | #4 of 10 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections