Learning to Compose Topic-Aware Mixture of Experts for Zero-Shot Video Captioning
This addresses the limitation of existing video captioning models that fail to generalize to open vocabulary scenarios, offering a solution for describing novel activities without paired training data.
The paper tackles the problem of zero-shot video captioning for out-of-domain videos with unseen activities, proposing a Topic-Aware Mixture of Experts model that transfers knowledge from seen to unseen activities, achieving strong generalization ability as validated empirically.
Although promising results have been achieved in video captioning, existing models are limited to the fixed inventory of activities in the training corpus, and do not generalize to open vocabulary scenarios. Here we introduce a novel task, zero-shot video captioning, that aims at describing out-of-domain videos of unseen activities. Videos of different activities usually require different captioning strategies in many aspects, i.e. word selection, semantic construction, and style expression etc, which poses a great challenge to depict novel activities without paired training data. But meanwhile, similar activities share some of those aspects in common. Therefore, We propose a principled Topic-Aware Mixture of Experts (TAMoE) model for zero-shot video captioning, which learns to compose different experts based on different topic embeddings, implicitly transferring the knowledge learned from seen activities to unseen ones. Besides, we leverage external topic-related text corpus to construct the topic embedding for each activity, which embodies the most relevant semantic vectors within the topic. Empirical results not only validate the effectiveness of our method in utilizing semantic knowledge for video captioning, but also show its strong generalization ability when describing novel activities.