VEATIC: Video-based Emotion and Affect Tracking in Context Dataset: Conclusion

cover
27 May 2024

Authors:

(1) Zhihang Ren, University of California, Berkeley and these authors contributed equally to this work (Email: peter.zhren@berkeley.edu);

(2) Jefferson Ortega, University of California, Berkeley and these authors contributed equally to this work (Email: jefferson_ortega@berkeley.edu);

(3) Yifan Wang, University of California, Berkeley and these authors contributed equally to this work (Email: wyf020803@berkeley.edu);

(4) Zhimin Chen, University of California, Berkeley (Email: zhimin@berkeley.edu);

(5) Yunhui Guo, University of Texas at Dallas (Email: yunhui.guo@utdallas.edu);

(6) Stella X. Yu, University of California, Berkeley and University of Michigan, Ann Arbor (Email: stellayu@umich.edu);

(7) David Whitney, University of California, Berkeley (Email: dwhitney@berkeley.edu).

6. Conclusion

In this study, we proposed the first context based large video dataset, VEATIC, for continuous valence and arousal prediction. Various visualizations show the diversity of our dataset and the consistency of our annotations. We also proposed a simple baseline algorithm to solve this challenge. Empirical results prove the effectiveness of our proposed method and the VEATIC dataset.

This paper is available on arxiv under CC 4.0 license.