Exploiting Feature Diversity for Make-up Temporal Video Grounding
This work addresses a domain-specific challenge in video understanding for make-up tutorials, but it is incremental as it builds on existing competition frameworks.
The paper tackled the problem of fine-grained video-text semantics for make-up step localization in untrimmed videos, achieving 3rd place in the MTVG competition by exploiting feature diversity through methods in feature extraction, network optimization, and model ensemble.
This technical report presents the 3rd winning solution for MTVG, a new task introduced in the 4-th Person in Context (PIC) Challenge at ACM MM 2022. MTVG aims at localizing the temporal boundary of the step in an untrimmed video based on a textual description. The biggest challenge of this task is the fi ne-grained video-text semantics of make-up steps. However, current methods mainly extract video features using action-based pre-trained models. As actions are more coarse-grained than make-up steps, action-based features are not sufficient to provide fi ne-grained cues. To address this issue,we propose to achieve fi ne-grained representation via exploiting feature diversities. Specifically, we proposed a series of methods from feature extraction, network optimization, to model ensemble. As a result, we achieved 3rd place in the MTVG competition.