Empirical autopsy of deep video captioning encoder-decoder architecture

Abstract

Contemporary deep learning based video captioning methods adopt encoder-decoder framework. In encoder, visual features are extracted with 2D/3D Convolutional Neural Networks (CNNs) and a transformed version of those features is passed to the decoder. The decoder uses word embeddings and a language model to map visual features to natural language captions. Due to its composite nature, the encoder-decoder pipeline provides the freedom of multiple choices for each of its components, e.g., the choices of CNN models, feature transformations, word embeddings, and language models etc. Component selection can have drastic effects on the overall video captioning performance. However, current literature is void of any systematic investigation in this regard. This article fills this gap by providing the first thorough empirical analysis of the role that each major component plays in a widely adopted video captioning pipeline. We perform extensive experiments by varying the constituent components of the video captioning framework, and quantify the performance gains that are possible by mere component selection. We use the popular MSVD dataset as the test-bed, and demonstrate that substantial performance gains are possible by careful selection of the constituent components without major changes to the pipeline itself. These results are expected to provide guiding principles for research in the fast growing direction of video captioning.

Keywords

• CNN architecture
• Encoder-decoder
• Word embeddings
• Language model
• Video captioning
• Language and vision