Hugging Face's logo

Datasets:
ANDRYHA
/
AudioVisualMouseSaliency

Size:
1K<n<10K
ArXiv:
License:
Dataset card Files Community
AudioVisualMouseSaliency / README.md
ANDRYHA's picture
ANDRYHA
Update README.md
9540aeb verified
preview code
|
raw
history blame
4.03 kB
metadata
license: cc-by-4.0
pretty_name: AViMoS
size_categories:
  - 1K<n<10K

ECCV-AIM Video Saliency Prediction Challenge 2024

Page Paper

Dataset

We provide a novel audio-visual mouse saliency (AViMoS) dataset with the following key-features:

  • Diverse content: movie, sports, live, vertical videos, etc.;
  • Large scale: 1500 videos with mean 19s duration;
  • High resolution: all streams are FullHD;
  • Audio track saved and played to observers;
  • Mouse fixations from >5000 observers (>70 per video);
  • License: CC-BY;

File structure:

  1. Videos.zip — 1500 (1000 Train + 500 Test) .mp4 video (kindly reminder: many videos contain an audio stream and users watched the video with the sound turned ON!)

  2. TrainTestSplit.json — in this JSON we provide Train/Public Test/Private Test split of all videos

  3. SaliencyTrain.zip/SaliencyTest.zip — almost losslessly (crf 0, 10bit, min-max normalized) compressed continuous saliency maps videos for Train/Test subset

  4. FixationsTrain.zip/FixationsTest.zip — contains the following files for Train/Test subset:

  • .../video_name/fixations.json — per-frame fixations coordinates, from which saliency maps were obtained, this JSON will be used for metrics calculation

  • .../video_name/fixations/ — binary fixation maps in '.png' format (since some fixations could share the same pixel, this is a lossy representation and is NOT used either in calculating metrics or generating Gaussians, however, we provide them for visualization and frames count checks)

  1. VideoInfo.json — meta information about each video (e.g. license)

Evaluation

Environment setup 

conda create -n saliency python=3.8.16
conda activate saliency
pip install numpy==1.24.2 opencv-python==4.7.0.72 tqdm==4.65.0
conda install ffmpeg=4.4.2 -c conda-forge

Run evaluation

Archives with videos were accepted from challenge participants as submissions and scored using the same pipeline as in bench.py.

Usage example:

  1. Check that your predictions match the structure and names of the baseline CenterPrior submission
  2. Install pip install -r requirments.txt, conda install ffmpeg
  3. Download and extract SaliencyTest.zip, FixationsTest.zip, and TrainTestSplit.json files from the dataset page
  4. Run python bench.py with flags:
  • --model_video_predictions ./SampleSubmission-CenterPrior — folder with predicted saliency videos
  • --model_extracted_frames ./SampleSubmission-CenterPrior-Frames — folder to store prediction frames (should not exist at launch time), requires ~170 GB of free space
  • --gt_video_predictions ./SaliencyTest/Test — folder from dataset page with gt saliency videos
  • --gt_extracted_frames ./SaliencyTest-Frames — folder to store ground-truth frames (should not exist at launch time), requires ~170 GB of free space
  • --gt_fixations_path ./FixationsTest/Test — folder from dataset page with gt saliency fixations
  • --split_json ./TrainTestSplit.json — JSON from dataset page with names splitting
  • --results_json ./results.json — path to the output results json
  • --mode public_test — public_test/private_test subsets
  1. The result you get will be available following results.json path

Citation

Please cite the paper if you find challenge materials useful for your research:

@article{ }

Challenges Benchmarks