Video-Based Sentiment Analysis with hvnLBP-TOP Feature and bi-LSTM

EasyChair Preprint 657

Abstract

In this paper, we propose a new feature extraction method called hvnLBP-TOP for video-based sentiment analysis. Furthermore, we use principal component analysis (PCA) and bidirectional long short term memory (bi-LSTM) for dimensionality reduction and classification. We achieved an average recognition accuracy of 71.1% on the MOUD dataset and 63.9% on the CMU-MOSI dataset.

Keyphrases: Sentiment Analysis, cmu mosi dataset, facial expression recognition, feature extraction, machine learning, moud dataset, video based sentiment analysis, video sentiment analysis

@booklet{EasyChair:657,
  author    = {Haoran Li and Hua Xu},
  title     = {Video-Based Sentiment Analysis with hvnLBP-TOP Feature and bi-LSTM},
  howpublished = {EasyChair Preprint 657},
  year      = {EasyChair, 2018}}