Camera Movement Classification in Historical Footage: A Comparative Study of Deep Video Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Tingyu, Dadras, Armin, Kleber, Florian, Sablatnig, Robert
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912652780896256
author Lin, Tingyu
Dadras, Armin
Kleber, Florian
Sablatnig, Robert
author_facet Lin, Tingyu
Dadras, Armin
Kleber, Florian
Sablatnig, Robert
contents Camera movement conveys spatial and narrative information essential for understanding video content. While recent camera movement classification (CMC) methods perform well on modern datasets, their generalization to historical footage remains unexplored. This paper presents the first systematic evaluation of deep video CMC models on archival film material. We summarize representative methods and datasets, highlighting differences in model design and label definitions. Five standard video classification models are assessed on the HISTORIAN dataset, which includes expert-annotated World War II footage. The best-performing model, Video Swin Transformer, achieves 80.25% accuracy, showing strong convergence despite limited training data. Our findings highlight the challenges and potential of adapting existing models to low-quality video and motivate future work combining diverse input modalities and temporal architectures.
format Preprint
id arxiv_https___arxiv_org_abs_2510_14713
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Camera Movement Classification in Historical Footage: A Comparative Study of Deep Video Models
Lin, Tingyu
Dadras, Armin
Kleber, Florian
Sablatnig, Robert
Computer Vision and Pattern Recognition
Artificial Intelligence
Image and Video Processing
Camera movement conveys spatial and narrative information essential for understanding video content. While recent camera movement classification (CMC) methods perform well on modern datasets, their generalization to historical footage remains unexplored. This paper presents the first systematic evaluation of deep video CMC models on archival film material. We summarize representative methods and datasets, highlighting differences in model design and label definitions. Five standard video classification models are assessed on the HISTORIAN dataset, which includes expert-annotated World War II footage. The best-performing model, Video Swin Transformer, achieves 80.25% accuracy, showing strong convergence despite limited training data. Our findings highlight the challenges and potential of adapting existing models to low-quality video and motivate future work combining diverse input modalities and temporal architectures.
title Camera Movement Classification in Historical Footage: A Comparative Study of Deep Video Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Image and Video Processing
url https://arxiv.org/abs/2510.14713