Abstract:
Multimodal summarization for open-domain videos is an emerging task, aiming to generate a summary from multisource information (video, audio, transcript). Despite the success of recent multiencoder-decoder frameworks on this task, existing methods lack fine-grained multimodality interactions of multisource inputs. Besides, unlike other multimodal tasks, this task has longer multimodal sequences with more redundancy and noise. To address these two issues, we propose a multistage fusion network with the fusion forget gate module, which builds upon this approach by modeling fine-grained interactions between the modalities through a multistep fusion schema and controlling the flow of redundant information between multimodal long sequences via a forgetting module. Experimental results on the How2 dataset show that our proposed model achieves a new state-of-the-art performance. Comprehensive analysis empirically verifies the effectiveness of our fusion schema and forgetting module on multiple encoder-decoder architectures. Specially, when using high noise ASR transcripts (WER\textgreater30%), our model still achieves performance close to the ground-truth transcript model, which reduces manual annotation cost.
