---
pretty_name: VGGSound-Duet^Mask
language:
- en
task_categories:
- image-segmentation
size_categories:
- 1K<n<10K
tags:
- audio-visual-localization
- sound-source-localization
- segmentation-masks
- vggsound
- multimodal
---

# VGGSound-Duet^Mask

**VGGSound-Duet^Mask** provides pixel-level segmentation masks for evaluating dual-source audio-visual localisation. It accompanies **SCAV**, introduced in *Whence the Voice? Self-supervised Dual-source Audio-Visual Localisation via Selective Convergence* (ECCV 2026).

For more information, visit the [project page](https://happy-new-bears.github.io/scav-project-page/), read the [paper](https://happy-new-bears.github.io/scav-project-page/Han_ECCV_2026_CAMERA_READY.pdf), or see the [official code](https://github.com/happy-new-bears/SCAV).

## Overview

Conventional sound-source localisation benchmarks frequently use bounding boxes, which include background pixels and can overestimate localisation quality. VGGSound-Duet^Mask replaces box-level evaluation with foreground segmentation masks so that predicted heatmaps can be compared with the visible sound-producing regions more precisely.

### Key features

- Pixel-level masks for the VGGSound-Duet benchmark.
- Designed for mask-based evaluation of dual-source audio-visual localisation.
- Supports the 3,951 currently accessible VGGSound-Duet test pairs spanning 220 sound categories described in the SCAV paper.
- Contains 4,512 individual source-frame mask files in the current Hugging Face repository. Individual masks are reused when constructing the dual-source test pairs.
- Masks were generated with the Segment Anything Model (SAM) and stored as NumPy arrays serialized with Python pickle.

This repository contains the **mask annotations only**. It does not redistribute the source videos, audio, extracted frames, optical flow, VGGSound metadata, or SCAV checkpoints.

## Download

The repository is public and does not require an access token.

### Hugging Face CLI

```bash
pip install -U "huggingface_hub[cli]"
hf download lemonweed6312/VGGSound-Duet-Mask \
  --repo-type dataset \
  --local-dir ./VGGSound-Duet-Mask
```

### Python

```python
from huggingface_hub import snapshot_download

snapshot_download(
    repo_id="lemonweed6312/VGGSound-Duet-Mask",
    repo_type="dataset",
    local_dir="./VGGSound-Duet-Mask",
)
```

To download one mask only:

```python
from huggingface_hub import hf_hub_download

mask_path = hf_hub_download(
    repo_id="lemonweed6312/VGGSound-Duet-Mask",
    repo_type="dataset",
    filename="masks/-0BIyqJj9ZU_000030.pkl",
)
```

## Dataset structure

```text
VGGSound-Duet-Mask/
├── README.md
└── masks/
    ├── -0BIyqJj9ZU_000030.pkl
    ├── -2-wdcN5vOw_000017.pkl
    └── ...
```

Each filename follows this pattern:

```text
<youtube_video_id>_<frame_index>.pkl
```

The frame index is zero-padded to six digits. Each pickle contains one NumPy mask array. The evaluation code accepts either a two-dimensional array `(H, W)` or a single-channel array `(H, W, 1)`. Spatial dimensions follow the corresponding source frame and may vary. Masks are interpreted as binary using a threshold of `0.5`.

For example, `-0BIyqJj9ZU_000030.pkl` contains a `float64` NumPy array of shape `(240, 320, 1)` with values `0` and `1`. Its size of approximately 615 kB is expected: an uncompressed `240 × 320 × 1` float64 array alone requires 614,400 bytes.

## Reading a mask

```python
import pickle
from pathlib import Path

import numpy as np

mask_path = Path("VGGSound-Duet-Mask/masks/-0BIyqJj9ZU_000030.pkl")

with mask_path.open("rb") as file:
    mask = pickle.load(file)

if mask.ndim == 3:
    mask = mask[:, :, 0]

mask = (mask > 0.5).astype(np.float32)
print(mask.shape, mask.dtype, mask.min(), mask.max())
```

The pretrained weights are available in the [SCAV model repository](https://huggingface.co/lemonweed6312/SCAV). See the [official code README](https://github.com/happy-new-bears/SCAV#evaluation-with-pretrained-weights) for the remaining data preparation and evaluation requirements.

## Dataset details

- **Developed by:** Han Hu, Dongheng Lin, Yuqi Hou, Haotian Li, Hyung Jin Chang, and Jianbo Jiao
- **Institution:** The MIx Group, University of Birmingham
- **Associated paper:** *Whence the Voice? Self-supervised Dual-source Audio-Visual Localisation via Selective Convergence*
- **Conference:** ECCV 2026
- **Task:** dual-source audio-visual localisation evaluation
- **Annotation type:** foreground segmentation mask
- **Number of mask files in this repository:** 4,512
- **Benchmark coverage reported in the paper:** 3,951 currently accessible dual-source test pairs across 220 sound categories

## Limitations and responsible use

- The masks are automatically generated with SAM and may contain segmentation errors, omissions, or ambiguous object boundaries.
- Dataset coverage depends on the availability of the original VGGSound videos; some source videos are no longer available on YouTube.
- The annotations identify visible foreground regions used for benchmark evaluation. They should not be interpreted as exhaustive semantic segmentation labels.
- Users are responsible for complying with the terms and licenses of VGGSound, YouTube, and any source media they obtain separately.

## Citation

If you use VGGSound-Duet^Mask or the SCAV checkpoints, please cite:

```bibtex
@inproceedings{hu2026scav,
  title     = {Whence the Voice? Self-supervised Dual-source Audio-Visual Localisation via Selective Convergence},
  author    = {Hu, Han and Lin, Dongheng and Hou, Yuqi and Li, Haotian and Chang, Hyung Jin and Jiao, Jianbo},
  booktitle = {European Conference on Computer Vision (ECCV)},
  year      = {2026}
}
```