Multi-Task Multi-Frame Visual Piano Transcription
ISMIR - 2026
Loading...
For fixed top-view keyboard video. Supply all four keyboard corners for perspective correction; V2N predicts physical key release and note velocity, not pedal-extended audio offsets.
Input
Supports inputs in formats such as mp4 / mov / mkv
Supported formats
Output
MIDI
Use case
Edit notes, instrumentation and performance details.
ISMIR - 2026
Includes public test results for models used here and elsewhere. The complete source table is shown by default; conditions and metrics can also be selected for comparison. Scores from different test conditions are not directly comparable.
Only records with verified public inference implementations and matching weights are shown; algorithms requiring no weights are also eligible. Closed services, unreleased weights and unmatched experiment versions are excluded. This is not an all-model leaderboard.
Public weights do not establish a fully open-source license. Expand the data for declared weight licenses and access conditions. Undeclared licenses are marked separately.
Choose a dataset to view all metrics for each model. Rows follow the official rank or selected metric. Bold marks each column’s best comparable value, including ties. Different evaluation protocols are ranked separately; missing scores are shown as -.
Deployable entries 1 / source rows 13
A complete public inference package for these experiments has not been verified. Other weights with the same model name do not qualify.
S2S † · Trained on PianoVAM → Test: PianoVAM · V2R † · Trained on PianoVAM → Test: PianoVAM · Li et al. † · Trained on PianoVAM → Test: PianoVAM · PPAN † · Trained on PianoVAM → Test: PianoVAM · S2S † · Trained on R3s+R3x → Test: R3s · V2R † · Trained on R3s+R3x → Test: R3s · PPAN † · Trained on R3s+R3x → Test: R3s · V2N (Ours) · Trained on R3s+R3x → Test: R3s · S2S † · Trained on R3s+R3x → Test: R3x · V2R † · Trained on R3s+R3x → Test: R3x · PPAN † · Trained on R3s+R3x → Test: R3x · V2N (Ours) · Trained on R3s+R3x → Test: R3x
Values come from the linked public source. Row labels identify the test set or checkpoint; unreported scores are shown as -.
| Metric rank | Model | Test setting | Frame · (60 Hz) | Onset | + Off | + Vel | + Off + Vel | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 50 ms↑ | 100 ms↑ | 50 ms↑ | 100 ms↑ | 50 ms↑ | 100 ms↑ | 50 ms↑ | 100 ms↑ | ||||
| - | V2N (Ours)Used on this sitePublic weights · Inference code | Trained on PianoVAM → Test: PianoVAM | 90.9 ∗ | 94.7 | 97.0 | 89.5 ∗ | 95.5 ∗ | 82.8 ∗ | 84.5 ∗ | 78.3 ∗ | 83.2 ∗ |
Markers identify models or components used here. Scores and settings come from the cited tests, not a new evaluation of the complete site pipeline.
Datasets, versions and conditions remain separate, including sources with no eligible ranking rows.
Deployable entries 1 / source rows 13
A complete public inference package for these experiments has not been verified. Other weights with the same model name do not qualify.
S2S † · Trained on PianoVAM → Test: PianoVAM · V2R † · Trained on PianoVAM → Test: PianoVAM · Li et al. † · Trained on PianoVAM → Test: PianoVAM · PPAN † · Trained on PianoVAM → Test: PianoVAM · S2S † · Trained on R3s+R3x → Test: R3s · V2R † · Trained on R3s+R3x → Test: R3s · PPAN † · Trained on R3s+R3x → Test: R3s · V2N (Ours) · Trained on R3s+R3x → Test: R3s · S2S † · Trained on R3s+R3x → Test: R3x · V2R † · Trained on R3s+R3x → Test: R3x · PPAN † · Trained on R3s+R3x → Test: R3x · V2N (Ours) · Trained on R3s+R3x → Test: R3x
Scores apply to the listed dataset, metric and model version; results from different tests are not directly comparable.
Choose a dataset to view all metrics for each model. Rows follow the official rank or selected metric. Bold marks each column’s best comparable value, including ties. Different evaluation protocols are ranked separately; missing scores are shown as -.
Deployable entries 1 / source rows 13
A complete public inference package for these experiments has not been verified. Other weights with the same model name do not qualify.
S2S † · Trained on PianoVAM → Test: PianoVAM · V2R † · Trained on PianoVAM → Test: PianoVAM · Li et al. † · Trained on PianoVAM → Test: PianoVAM · PPAN † · Trained on PianoVAM → Test: PianoVAM · S2S † · Trained on R3s+R3x → Test: R3s · V2R † · Trained on R3s+R3x → Test: R3s · PPAN † · Trained on R3s+R3x → Test: R3s · V2N (Ours) · Trained on R3s+R3x → Test: R3s · S2S † · Trained on R3s+R3x → Test: R3x · V2R † · Trained on R3s+R3x → Test: R3x · PPAN † · Trained on R3s+R3x → Test: R3x · V2N (Ours) · Trained on R3s+R3x → Test: R3x
Values come from the linked public source. Row labels identify the test set or checkpoint; unreported scores are shown as -.
| Metric rank | Model | Test setting | Frame · (60 Hz) | Onset | + Off | + Vel | + Off + Vel | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 50 ms↑ | 100 ms↑ | 50 ms↑ | 100 ms↑ | 50 ms↑ | 100 ms↑ | 50 ms↑ | 100 ms↑ | ||||
| - | V2N (Ours)Public weights · Inference code | Trained on PianoVAM → Test: PianoVAM | 90.9 ∗ | 94.7 | 97.0 | 89.5 ∗ | 95.5 ∗ | 82.8 ∗ | 84.5 ∗ | 78.3 ∗ | 83.2 ∗ |
Include the tool name, steps and the problem in your feedback. Screenshots can help us locate it.