LookAt-Data

This repository contains the motion capture data pertraining to our work Generating Emotionally Expressive Look-At Animation.

File Naming

Files are named in the format [RecordingNumber]_[EngagingStyle][EngagingStyleIntensity]_[DisengagingStyle][DisengagingStyleIntensity]. The style names are herby abbreviated to a single letter based on the following mapping:

• C = Confrontational
• D = Down
• F = Fearful
• G = Disgust
• H = Happy
• I = Interested
• S = Surprise
• N1 = Bored
• N2 = Neutral
• N3 = Distracted

Each style except the 'N' styles (neutral, bored, distracted) was captured with 2 intensity, for example, 'C2' and 'C3' refer to medium and high level expression of confrontation, respectively.

When recoding of a style combination was repeated, _000 and _001 are added to the file name for the first and second repeat, respectively.

File Contents

Each file contains the full body motion-capture plus the tracked motion of the two look-at targets. The actor alternates between the two targets, which are moved after each look-at take.
Start frame of each look-at take at 120fps are listed in all_frame_ranges.csv. End frames are simply listed as the start frame of the next take.

Using this data

This dataset is for non-commercial research purposes only. If you use this dataset, you must cite the paper:

@inproceedings{ferstl2023generating,
  title={Generating Emotionally Expressive Look-At Animation},
  author={Ferstl, Ylva},
  booktitle={Proceedings of the 16th ACM SIGGRAPH Conference on Motion, Interaction and Games},
  pages={1--6},
  year={2023}
}

© [2023] Ubisoft Entertainment. All Rights Reserved