Crowdsourcing More Effective Initializations for Single-Target Trackers Through Automatic Re-querying

要旨

In single-target video object tracking, an initial bounding box is drawn around a target object and propagated through a video. When this bounding box is provided by a careful human expert, it is expected to yield strong overall tracking performance that can be mimicked at scale by novice crowd workers with the help of advanced quality control methods. However, we show through an investigation of 900 crowdsourced initializations that such quality control strategies are inadequate for this task in two major ways: first, the high level of redundancy in these methods (e.g., averaging multiple responses to reduce error) is unnecessary, as 23\% of crowdsourced initializations perform just as well as the gold-standard initialization. Second, even nearly perfect initializations can lead to degraded long-term performance due to the complexity of object tracking. Considering these findings, we evaluate novel approaches for automatically selecting bounding boxes to re-query, and introduce \textit{Smart Replacement}, an efficient method that decides whether to use the crowdsourced replacement initialization.

著者
Stephan J. Lemmer
University of Michigan, Ann Arbor, Michigan, United States
Jean Y. Song
KAIST, Daejeon, Korea, Republic of
Jason J. Corso
Stevens Institute for Artificial Intelligence, Hoboken, New Jersey, United States
DOI

10.1145/3411764.3445181

論文URL

https://doi.org/10.1145/3411764.3445181

動画

会議: CHI 2021

The ACM CHI Conference on Human Factors in Computing Systems (https://chi2021.acm.org/)

セッション: Computational Human-AI Conversation

[A] Paper Room 02, 2021-05-11 17:00:00~2021-05-11 19:00:00 / [B] Paper Room 02, 2021-05-12 01:00:00~2021-05-12 03:00:00 / [C] Paper Room 02, 2021-05-12 09:00:00~2021-05-12 11:00:00
Paper Room 02
14 件の発表
2021-05-11 17:00:00
2021-05-11 19:00:00
日本語まとめ
読み込み中…