Who's Better? Who's Best? Pairwise Deep Ranking for Skill Determination
This work addresses automated organization of how-to video collections and generic skill determination, but it is incremental as it builds on existing deep ranking methods with a novel loss function.
The paper tackles the problem of assessing skill from video across tasks like surgery and drawing by formulating it as pairwise and overall ranking, achieving 70% to 83% correct ordering of video pairs on four datasets.
We present a method for assessing skill from video, applicable to a variety of tasks, ranging from surgery to drawing and rolling pizza dough. We formulate the problem as pairwise (who's better?) and overall (who's best?) ranking of video collections, using supervised deep ranking. We propose a novel loss function that learns discriminative features when a pair of videos exhibit variance in skill, and learns shared features when a pair of videos exhibit comparable skill levels. Results demonstrate our method is applicable across tasks, with the percentage of correctly ordered pairs of videos ranging from 70% to 83% for four datasets. We demonstrate the robustness of our approach via sensitivity analysis of its parameters. We see this work as effort toward the automated organization of how-to video collections and overall, generic skill determination in video.