Sequence to Sequence -- Video to Text
This addresses the challenge of describing complex video dynamics for applications in video understanding and accessibility, though it is incremental as it builds on existing LSTM methods from image captioning.
The authors tackled the problem of generating open-domain video descriptions by proposing an end-to-end sequence-to-sequence model using LSTMs, which learns to associate video frames with words and achieves performance on standard datasets like YouTube, M-VAD, and MPII-MD.
Real-world videos often have complex dynamics; and methods for generating open-domain video descriptions should be sensitive to temporal structure and allow both input (sequence of frames) and output (sequence of words) of variable length. To approach this problem, we propose a novel end-to-end sequence-to-sequence model to generate captions for videos. For this we exploit recurrent neural networks, specifically LSTMs, which have demonstrated state-of-the-art performance in image caption generation. Our LSTM model is trained on video-sentence pairs and learns to associate a sequence of video frames to a sequence of words in order to generate a description of the event in the video clip. Our model naturally is able to learn the temporal structure of the sequence of frames as well as the sequence model of the generated sentences, i.e. a language model. We evaluate several variants of our model that exploit different visual features on a standard set of YouTube videos and two movie description datasets (M-VAD and MPII-MD).