Deep Learning Based Semantic Video Indexing and Retrieval

TL;DR

This paper presents a video retrieval system leveraging deep convolutional neural network features and a graph-based indexing structure to enable efficient semantic search with complex spatial and temporal queries.

Contribution

It introduces a novel approach combining CNN-extracted features with graph-based storage for improved semantic video retrieval.

Findings

Deep learned features serve as universal signatures for video content.

Graph-based indexing enables efficient retrieval of complex spatial and temporal queries.

The system demonstrates effective performance in semantic video search tasks.

Abstract

We share the implementation details and testing results for video retrieval system based exclusively on features extracted by convolutional neural networks. We show that deep learned features might serve as universal signature for semantic content of video useful in many search and retrieval tasks. We further show that graph-based storage structure for video index allows to efficiently retrieving the content with complicated spatial and temporal search queries.