End-to-End Generative Pretraining for Multimodal Video Captioning

Paul Hongsuck Seo, Arsha Nagrani, Anurag Arnab, Cordelia Schmid; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 17959-17968

Abstract


Recent video and language pretraining frameworks lack the ability to generate sentences. We present Multimodal Video Generative Pretraining (MV-GPT), a new pretraining framework for learning from unlabelled videos which can be effectively used for generative tasks such as multimodal video captioning. Unlike recent video-language pretraining frameworks, our framework trains both a multimodal video encoder and a sentence decoder jointly. To overcome the lack of captions in unlabelled videos, we leverage the future utterance as an additional text source and propose a bidirectional generation objective -- we generate future utterances given the present mulitmodal context, and also the present utterance given future observations. With this objective, we train an encoder-decoder model end-to-end to generate a caption from raw pixels and transcribed speech directly. Our model achieves state-of-the-art performance for multimodal video captioning on four standard benchmarks, as well as for other video understanding tasks such as generative and discriminative VideoQA, video retrieval and action classification.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Seo_2022_CVPR, author = {Seo, Paul Hongsuck and Nagrani, Arsha and Arnab, Anurag and Schmid, Cordelia}, title = {End-to-End Generative Pretraining for Multimodal Video Captioning}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2022}, pages = {17959-17968} }