# Video Infringement Detection via Feature Disentanglement and Mutual Information Maximization

arXiv: https://arxiv.org/abs/2309.06877 · HTML: https://mlsanity.com/paper/2309.06877/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 45 references that are in the map.
- First posted on arXiv 13 Sep 2023.

## Builds on (most cited first, 23 of 23)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [An Empirical Study of Training Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.02057/) (2021)
- [Deep Learning and the Information Bottleneck Principle](https://mlsanity.com/paper/1503.02406/) (2015)
- [Particular object retrieval with integral max-pooling of CNN activations](https://mlsanity.com/paper/1511.05879/) (2015)
- [Predicting Visual Features from Text for Image and Video Caption Retrieval](https://mlsanity.com/paper/1709.01362/) (2017)
- [Deep Dual Consecutive Network for Human Pose Estimation](https://mlsanity.com/paper/2103.07254/) (2021)
- [Farewell to Mutual Information: Variational Distillation for Cross-Modal Person Re-Identification](https://mlsanity.com/paper/2104.02862/) (2021)
- [RGB-D Saliency Detection via Cascaded Mutual Information Minimization](https://mlsanity.com/paper/2109.07246/) (2021)
- [Tree-Augmented Cross-Modal Encoding for Complex-Query Video Retrieval](https://mlsanity.com/paper/2007.02503/) (2020)
- [ViSiL: Fine-grained Spatio-Temporal Video Similarity Learning](https://mlsanity.com/paper/1908.07410/) (2019)
- [Temporal Feature Alignment and Mutual Information Maximization for Video-Based Human Pose Estimation](https://mlsanity.com/paper/2203.15227/) (2022)
- [Contrastive Learning with Large Memory Bank and Negative Embedding Subtraction for Accurate Copy Detection](https://mlsanity.com/paper/2112.04323/) (2021)
- [Copy Motion From One to Another: Fake Motion Video Generation](https://mlsanity.com/paper/2205.01373/) (2022)
- [Attention in Attention: Modeling Context Correlation for Efficient Video Classification](https://mlsanity.com/paper/2204.09303/) (2022)
- [A Large-scale Comprehensive Dataset and Copy-overlap Aware Evaluation Protocol for Segment-level Video Copy Detection](https://mlsanity.com/paper/2203.02654/) (2022)
- [Action Recognition with Multi-stream Motion Modeling and Mutual Information Maximization](https://mlsanity.com/paper/2306.07576/) (2023)
- [Self-supervised Video Retrieval Transformer Network](https://mlsanity.com/paper/2104.07993/) (2021)

## Abstract (arXiv)

The self-media era provides us tremendous high quality videos. Unfortunately, frequent video copyright infringements are now seriously damaging the interests and enthusiasm of video creators. Identifying infringing videos is therefore a compelling task. Current state-of-the-art methods tend to simply feed high-dimensional mixed video features into deep neural networks and count on the networks to extract useful representations. Despite its simplicity, this paradigm heavily relies on the original entangled features and lacks constraints guaranteeing that useful task-relevant semantics are extracted from the features. In this paper, we seek to tackle the above challenges from two aspects: (1) We propose to disentangle an original high-dimensional feature into multiple sub-features, explicitly disentangling the feature into exclusive lower-dimensional components. We expect the sub-features to encode non-overlapping semantics of the original feature and remove redundant information. (2) On top of the disentangled sub-features, we further learn an auxiliary feature to enhance the sub-features. We theoretically analyzed the mutual information between the label and the disentangled features, arriving at a loss that maximizes the extraction of task-relevant information from the original feature. Extensive experiments on two large-scale benchmark datasets (i.e., SVD and VCSL) demonstrate that our method achieves 90.1% TOP-100 mAP on the large-scale SVD dataset and also sets the new state-of-the-art on the VCSL benchmark dataset. Our code and model have been released at https://github.com/yyyooooo/DMI/, hoping to contribute to the community.