VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modeling
Paper • 2406.04321 • Published • 1
Datasets:
Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
samples: list<item: struct<_id: struct<$oid: string>, filepath: string, tags: list<item: string>, _media_type (... 327 chars omitted)
child 0, item: struct<_id: struct<$oid: string>, filepath: string, tags: list<item: string>, _media_type: string, _ (... 315 chars omitted)
child 0, _id: struct<$oid: string>
child 0, $oid: string
child 1, filepath: string
child 2, tags: list<item: string>
child 0, item: string
child 3, _media_type: string
child 4, _rand: double
child 5, youtube_id: string
child 6, subset: string
child 7, duration_s: double
child 8, message_count: int64
child 9, channel_count: int64
child 10, topics: list<item: string>
child 0, item: string
child 11, schemas: list<item: string>
child 0, item: string
child 12, has_video: bool
child 13, has_audio: bool
child 14, _dataset_id: struct<$oid: string>
child 0, $oid: string
child 15, created_at: struct<$date: string>
child 0, $date: string
child 16, last_modified_at: struct<$date: string>
child 0, $date: string
group_media_types: null
app_config: null
classes: null
mask_targets: null
default_mask_targets: null
skeletons: null
camera_intrinsics: null
static_transforms: null
annotation_runs: null
brain_methods: null
evaluations: null
runs: null
label_schemas: null
frame_label_schemas: null
sample_fields: list<item: struct<name: string, ftype: string,
...
child 0, $date: string
child 5, db_field: string
child 6, description: null
child 7, info: null
child 8, read_only: bool
child 9, created_at: struct<$date: string>
child 0, $date: string
_id: struct<$oid: string>
child 0, $oid: string
default_classes: list<item: null>
child 0, item: null
_media_roots: list<item: null>
child 0, item: null
last_modified_at: struct<$date: string>
child 0, $date: string
created_at: struct<$date: string>
child 0, $date: string
_media_source_layouts: list<item: null>
child 0, item: null
media_type: string
saved_views: list<item: null>
child 0, item: null
sample_collection_name: string
info: struct<source: string, paper: string, license: string, subset_id_counts: struct<V2M: int64, V2M-20k: (... 40 chars omitted)
child 0, source: string
child 1, paper: string
child 2, license: string
child 3, subset_id_counts: struct<V2M: int64, V2M-20k: int64, V2M-bench: int64>
child 0, V2M: int64
child 1, V2M-20k: int64
child 2, V2M-bench: int64
child 4, note: string
_media_sources: list<item: null>
child 0, item: null
frame_fields: list<item: null>
child 0, item: null
last_loaded_at: struct<$date: string>
child 0, $date: string
persistent: bool
active_label_schemas: list<item: null>
child 0, item: null
tags: list<item: null>
child 0, item: null
slug: string
version: string
name: string
last_deletion_at: null
workspaces: list<item: null>
child 0, item: null
to
{'_id': {'$oid': Value('string')}, 'name': Value('string'), 'slug': Value('string'), 'version': Value('string'), 'created_at': {'$date': Value('string')}, 'last_modified_at': {'$date': Value('string')}, 'last_deletion_at': Value('null'), 'last_loaded_at': {'$date': Value('string')}, 'sample_collection_name': Value('string'), 'persistent': Value('bool'), 'media_type': Value('string'), '_media_roots': List(Value('null')), '_media_sources': List(Value('null')), '_media_source_layouts': List(Value('null')), 'group_media_types': Json(decode=True), 'tags': List(Value('null')), 'info': {'source': Value('string'), 'paper': Value('string'), 'license': Value('string'), 'subset_id_counts': {'V2M': Value('int64'), 'V2M-20k': Value('int64'), 'V2M-bench': Value('int64')}, 'note': Value('string')}, 'app_config': {'dynamic_groups_target_frame_rate': Value('int64'), 'grid_media_field': Value('string'), 'media_fallback': Value('bool'), 'media_fields': List(Value('string')), 'modal_media_field': Value('string'), 'plugins': Json(decode=True)}, 'classes': Json(decode=True), 'default_classes': List(Value('null')), 'mask_targets': Json(decode=True), 'default_mask_targets': Json(decode=True), 'skeletons': Json(decode=True), 'camera_intrinsics': Json(decode=True), 'static_transforms': Json(decode=True), 'sample_fields': List({'name': Value('string'), 'ftype': Value('string'), 'embedded_doc_type': Value('string'), 'subfield': Value('string'), 'fields': List({'name': Value('string'), 'ftype': Value('string'), 'embedded_doc_type': Value('null'), 'subfield': Value('null'), 'fields': List(Value('null')), 'db_field': Value('string'), 'description': Value('null'), 'info': Value('null'), 'read_only': Value('bool'), 'created_at': {'$date': Value('string')}}), 'db_field': Value('string'), 'description': Value('null'), 'info': Value('null'), 'read_only': Value('bool'), 'created_at': {'$date': Value('string')}}), 'frame_fields': List(Value('null')), 'saved_views': List(Value('null')), 'workspaces': List(Value('null')), 'annotation_runs': Json(decode=True), 'brain_methods': Json(decode=True), 'evaluations': Json(decode=True), 'runs': Json(decode=True), 'active_label_schemas': List(Value('null')), 'label_schemas': Json(decode=True), 'frame_label_schemas': Json(decode=True)}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2951, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2461, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2486, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 547, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 430, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
samples: list<item: struct<_id: struct<$oid: string>, filepath: string, tags: list<item: string>, _media_type (... 327 chars omitted)
child 0, item: struct<_id: struct<$oid: string>, filepath: string, tags: list<item: string>, _media_type: string, _ (... 315 chars omitted)
child 0, _id: struct<$oid: string>
child 0, $oid: string
child 1, filepath: string
child 2, tags: list<item: string>
child 0, item: string
child 3, _media_type: string
child 4, _rand: double
child 5, youtube_id: string
child 6, subset: string
child 7, duration_s: double
child 8, message_count: int64
child 9, channel_count: int64
child 10, topics: list<item: string>
child 0, item: string
child 11, schemas: list<item: string>
child 0, item: string
child 12, has_video: bool
child 13, has_audio: bool
child 14, _dataset_id: struct<$oid: string>
child 0, $oid: string
child 15, created_at: struct<$date: string>
child 0, $date: string
child 16, last_modified_at: struct<$date: string>
child 0, $date: string
group_media_types: null
app_config: null
classes: null
mask_targets: null
default_mask_targets: null
skeletons: null
camera_intrinsics: null
static_transforms: null
annotation_runs: null
brain_methods: null
evaluations: null
runs: null
label_schemas: null
frame_label_schemas: null
sample_fields: list<item: struct<name: string, ftype: string,
...
child 0, $date: string
child 5, db_field: string
child 6, description: null
child 7, info: null
child 8, read_only: bool
child 9, created_at: struct<$date: string>
child 0, $date: string
_id: struct<$oid: string>
child 0, $oid: string
default_classes: list<item: null>
child 0, item: null
_media_roots: list<item: null>
child 0, item: null
last_modified_at: struct<$date: string>
child 0, $date: string
created_at: struct<$date: string>
child 0, $date: string
_media_source_layouts: list<item: null>
child 0, item: null
media_type: string
saved_views: list<item: null>
child 0, item: null
sample_collection_name: string
info: struct<source: string, paper: string, license: string, subset_id_counts: struct<V2M: int64, V2M-20k: (... 40 chars omitted)
child 0, source: string
child 1, paper: string
child 2, license: string
child 3, subset_id_counts: struct<V2M: int64, V2M-20k: int64, V2M-bench: int64>
child 0, V2M: int64
child 1, V2M-20k: int64
child 2, V2M-bench: int64
child 4, note: string
_media_sources: list<item: null>
child 0, item: null
frame_fields: list<item: null>
child 0, item: null
last_loaded_at: struct<$date: string>
child 0, $date: string
persistent: bool
active_label_schemas: list<item: null>
child 0, item: null
tags: list<item: null>
child 0, item: null
slug: string
version: string
name: string
last_deletion_at: null
workspaces: list<item: null>
child 0, item: null
to
{'_id': {'$oid': Value('string')}, 'name': Value('string'), 'slug': Value('string'), 'version': Value('string'), 'created_at': {'$date': Value('string')}, 'last_modified_at': {'$date': Value('string')}, 'last_deletion_at': Value('null'), 'last_loaded_at': {'$date': Value('string')}, 'sample_collection_name': Value('string'), 'persistent': Value('bool'), 'media_type': Value('string'), '_media_roots': List(Value('null')), '_media_sources': List(Value('null')), '_media_source_layouts': List(Value('null')), 'group_media_types': Json(decode=True), 'tags': List(Value('null')), 'info': {'source': Value('string'), 'paper': Value('string'), 'license': Value('string'), 'subset_id_counts': {'V2M': Value('int64'), 'V2M-20k': Value('int64'), 'V2M-bench': Value('int64')}, 'note': Value('string')}, 'app_config': {'dynamic_groups_target_frame_rate': Value('int64'), 'grid_media_field': Value('string'), 'media_fallback': Value('bool'), 'media_fields': List(Value('string')), 'modal_media_field': Value('string'), 'plugins': Json(decode=True)}, 'classes': Json(decode=True), 'default_classes': List(Value('null')), 'mask_targets': Json(decode=True), 'default_mask_targets': Json(decode=True), 'skeletons': Json(decode=True), 'camera_intrinsics': Json(decode=True), 'static_transforms': Json(decode=True), 'sample_fields': List({'name': Value('string'), 'ftype': Value('string'), 'embedded_doc_type': Value('string'), 'subfield': Value('string'), 'fields': List({'name': Value('string'), 'ftype': Value('string'), 'embedded_doc_type': Value('null'), 'subfield': Value('null'), 'fields': List(Value('null')), 'db_field': Value('string'), 'description': Value('null'), 'info': Value('null'), 'read_only': Value('bool'), 'created_at': {'$date': Value('string')}}), 'db_field': Value('string'), 'description': Value('null'), 'info': Value('null'), 'read_only': Value('bool'), 'created_at': {'$date': Value('string')}}), 'frame_fields': List(Value('null')), 'saved_views': List(Value('null')), 'workspaces': List(Value('null')), 'annotation_runs': Json(decode=True), 'brain_methods': Json(decode=True), 'evaluations': Json(decode=True), 'runs': Json(decode=True), 'active_label_schemas': List(Value('null')), 'label_schemas': Json(decode=True), 'frame_label_schemas': Json(decode=True)}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Dataset Card for VidMuse-V2M-MCAP
This is a FiftyOne multimodal dataset with 300 samples. Each sample is one .mcap episode holding a video stream and its paired music stream.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from huggingface_hub import snapshot_download
# Download the dataset snapshot to the current working directory
snapshot_download(
repo_id="harpreetsahota/VidMuse-V2M-MCAP",
local_dir=".",
repo_type="dataset",
)
# Load dataset from current directory using FiftyOne's native format
dataset = fo.Dataset.from_dir(
dataset_dir=".", # Current directory contains the dataset files
dataset_type=fo.types.FiftyOneDataset, # Specify FiftyOne dataset format
name="VidMuse-V2M-MCAP", # Assign a name to the dataset for identification
)
# Launch the App
session = fo.launch_app(dataset)
Dataset Details
Dataset Description
V2M is a video-to-music dataset proposed in the VidMuse project, built from YouTube videos of genres such as movie trailers, advertisements, documentaries and vlogs. The paper describes three subsets: V2M (360K video-music pairs, ~1.8x10^4 hours, for pretraining), V2M-20K (20K pairs, ~6x10^2 hours, for finetuning) and V2M-bench (300 pairs, 9 hours, for evaluation). The source Hugging Face repo ships media only for V2M-bench; V2M and V2M-20K are provided as lists of YouTube IDs. This FiftyOne dataset is built from the V2M-bench media, with each video-music pair authored into a single MCAP file so the FiftyOne App can play video and music together.
- Curated by: Zeyue Tian, Zhaoyang Liu, Ruibin Yuan, Jiahao Pan, Qifeng Liu, Xu Tan, Qifeng Chen, Wei Xue, Yike Guo (HKUST Audio)
- Funded by: [More Information Needed]
- Shared by: [More Information Needed]
- Language(s): en
- License: CC BY-NC 4.0
Dataset Sources
- Repository: https://huggingface.co/datasets/HKUSTAudio/VidMuse-V2M-Dataset
- Paper: https://arxiv.org/abs/2406.04321
- Demo: [More Information Needed]
Uses
Direct Use
Research on video-to-music generation, including benchmarking models on V2M-bench.
Out-of-Scope Use
The data is licensed CC BY-NC 4.0, so commercial use is not permitted. [More Information Needed] for other out-of-scope uses.
Dataset Structure
This is a FiftyOne multimodal dataset (media_type="multimodal") with 300 samples. Each sample is one episode: a single .mcap file named {youtube_id}.mcap, 30 to 380 seconds long (about 9.0 hours in total, 9.2 GB). Every episode has two channels:
| Topic | MCAP schema | Content |
|---|---|---|
/video |
foxglove.CompressedVideo |
H.264 video (Main profile, level 3.1, no B-frames, SPS/PPS repeated on every keyframe), one frame per message, 720 px tall; 297 episodes are 1280x720, the other three are 1194x720, 1274x720 and 1278x720 (source resolution is preserved) |
/music |
foxglove.RawAudio |
Paired music track, pcm-s16, 44.1 kHz, stereo, one-second chunks per message |
Timestamps are offsets from the start of the episode (both streams start at 0), because the source release has no capture timestamps.
| Field | FiftyOne type | Description |
|---|---|---|
filepath |
StringField |
Path to the episode .mcap file |
tags |
ListField(StringField) |
bench on all samples |
metadata |
EmbeddedDocumentField |
FiftyOne sample metadata |
youtube_id |
StringField |
YouTube video ID, the filename stem of the source video and music files |
subset |
StringField |
V2M subset name; always V2M-bench here |
duration_s |
FloatField |
Episode duration in seconds, read from the MCAP summary |
message_count |
IntField |
Total messages in the MCAP |
channel_count |
IntField |
Number of channels (2 for every episode) |
topics |
ListField(StringField) |
Topic names in the file |
schemas |
ListField(StringField) |
MCAP schema names in the file |
has_video |
BooleanField |
True when a video/image schema is present (true for all 300) |
has_audio |
BooleanField |
True when an audio schema is present (true for all 300) |
There are no per-sample labels in the source release (no captions or genre labels), so the dataset has no label fields.
dataset.info: source repo URL, paper URL, license, subset_id_counts (V2M: 359,319 IDs, V2M-20k: 20,000 IDs, V2M-bench: 300 IDs) and a note that media is only available for V2M-bench.
Parsing decisions:
- Source files are
benchmark/video/{id}.mp4(silent video, no audio stream) andbenchmark/music/{id}.wav. They are paired by filename stem; all 300 stems match each other and the IDs inV2M-bench.txt. - The source videos were re-encoded to H.264 without B-frames (x264,
veryfast, CRF 26, GOP 30) because the FiftyOne multimodal viewer requires B-frame-free H.264, so the video in this dataset is a lossy re-encode of the original. The wav audio is stored unmodified as 16-bit PCM. - The V2M-20k and V2M lists are ID lists with no media, so only their counts are recorded in
dataset.info. The V2M-bench IDs are disjoint from both lists. - Checks run on every episode: MCAP file integrity; audio sample count equal to the source wav; decoded video frame count and dimensions equal to the source video; first video packet is a keyframe with SPS/PPS; decode order equals presentation order. Video and audio end within 0.10 s of each other (mean 0.05 s).
Dataset Creation
Curation Rationale
Existing video-music pair datasets were limited in scale, quality or genre diversity, so the authors built a large-scale dataset together with a benchmark for video-to-music generation.
Source Data
Data Collection and Processing
Video-music pairs were collected from YouTube, including data from existing resources such as HIMV-200K and YouTube-8M subsets tagged "Music" and "Trailer". The authors then applied a multi-step pipeline:
- Coarse filtering: removed videos missing audio or video tracks, videos that were too short or too long, inappropriate content such as violence or explicit material, and categories such as Interview, News and Gaming.
- Fine-grained filtering: kept videos with substantial music content (PANNs sound event detection; at least 50% of audio frames classified as music at a 0.5 confidence threshold) and removed videos made up mainly of static images (SSIM threshold of 0.8).
- Music source separation: isolated the music component by removing speech and singing.
- Audio-visual alignment ranking: ranked videos by audio-visual alignment score to keep strongly correlated pairs, then split the data into V2M, V2M-20K and V2M-bench subsets.
Human experts curated the benchmark subset.
Who are the source data producers?
The creators and uploaders of the YouTube videos. [More Information Needed] on any further demographic details.
Annotations
Annotation process
The source release contains no per-sample annotations. The V2M-bench subset was curated by human experts from the filtered data; the paper does not describe further annotation.
Who are the annotators?
Human experts curated the benchmark subset. [More Information Needed] for further details.
Personal and Sensitive Information
[More Information Needed]
Citation
BibTeX:
@article{tian2024vidmuse,
title={Vidmuse: A simple video-to-music generation framework with long-short-term modeling},
author={Tian, Zeyue and Liu, Zhaoyang and Yuan, Ruibin and Pan, Jiahao and Liu, Qifeng and Tan, Xu and Chen, Qifeng and Xue, Wei and Guo, Yike},
journal={arXiv preprint arXiv:2406.04321},
year={2024}
}
APA:
Tian, Z., Liu, Z., Yuan, R., Pan, J., Liu, Q., Tan, X., Chen, Q., Xue, W., & Guo, Y. (2024). VidMuse: A simple video-to-music generation framework with long-short-term modeling. arXiv preprint arXiv:2406.04321.
More Information
The source repo lists three ID files: V2M.txt, V2M-20k.txt and V2M-bench.txt. A grouped-slice version of this same data (silent video plus a .wav slice) is published as harpreetsahota/VidMuse-V2M.
Dataset Card Authors
[More Information Needed]
Dataset Card Contact
[More Information Needed]
- Downloads last month
- 332