About this document
Deep Learning for Video Summarization by Hoang Lại is a document available to read on EtoBox.
The document presents a project on deep reinforcement learning for unsupervised video summarization, focusing on creating concise, diverse, and representative video summaries. It outlines the development of a Deep Summarization Network (DSN) using CNNs and BiLSTM networks, along with a reinforcement learning approach to optimize video summarization. The project also includes a Kernel Temporal Segmentation (KTS) method for segmenting videos into visually or semantically consistent parts.
- Author
- Hoang Lại
- Language
- EN