# VideoMix: Rethinking Data Augmentation for Video Classification

arXiv: https://arxiv.org/abs/2012.03457 · HTML: https://mlsanity.com/paper/2012.03457/

## Key facts

- Cited by 55 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Dec 2020.

## Built on by (most cited first, 30 of 55)

- [VideoLT: Large-scale Long-tailed Video Recognition](https://mlsanity.com/paper/2105.02668/) (2021)
- [AugLy: Data Augmentations for Robustness](https://mlsanity.com/paper/2201.06494/) (2022)
- [A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainability](https://mlsanity.com/paper/2212.10888/) (2022)
- [Learn2Augment: Learning to Composite Videos for Data Augmentation in Action Recognition](https://mlsanity.com/paper/2206.04790/) (2022)
- [A Unified Analysis of Mixed Sample Data Augmentation: A Loss Function Perspective](https://mlsanity.com/paper/2208.09913/) (2022)
- [Mitigating and Evaluating Static Bias of Action Representations in the Background and the Foreground](https://mlsanity.com/paper/2211.12883/) (2022)
- [Green Edge AI: A Contemporary Survey](https://mlsanity.com/paper/2312.00333/) (2023)
- [Learning Representational Invariances for Data-Efficient Action Recognition](https://mlsanity.com/paper/2103.16565/) (2021)
- [ViewCLR: Learning Self-supervised Video Representation for Unseen Viewpoints](https://mlsanity.com/paper/2112.03905/) (2021)
- [VISTA: Enhancing Long-Duration and High-Resolution Video Understanding by Video Spatiotemporal Augmentation](https://mlsanity.com/paper/2412.00927/) (2024)
- [A Feature-space Multimodal Data Augmentation Technique for Text-video Retrieval](https://mlsanity.com/paper/2208.02080/) (2022)
- [Video OWL-ViT: Temporally-consistent open-world localization in video](https://mlsanity.com/paper/2308.11093/) (2023)
- [Benchmarking the Robustness of Temporal Action Detection Models Against Temporal Corruptions](https://mlsanity.com/paper/2403.20254/) (2024)
- [Exploring Temporally Dynamic Data Augmentation for Video Recognition](https://mlsanity.com/paper/2206.15015/) (2022)
- [A Unified Multimodal De- and Re-coupling Framework for RGB-D Motion Recognition](https://mlsanity.com/paper/2211.09146/) (2022)
- [A Simple Background Augmentation Method for Object Detection with Diffusion Model](https://mlsanity.com/paper/2408.00350/) (2024)
- [ObjectMix: Data Augmentation by Copy-Pasting Objects in Videos for Action Recognition](https://mlsanity.com/paper/2204.00239/) (2022)
- [An Efficient Framework for Few-shot Skeleton-based Temporal Action Segmentation](https://mlsanity.com/paper/2207.09925/) (2022)
- [An Empirical Study and Analysis on Open-Set Semi-Supervised Learning](https://mlsanity.com/paper/2101.08237/) (2021)
- [Out-Of-Distribution Detection with Diversification (Provably)](https://mlsanity.com/paper/2411.14049/) (2024)
- [Learning Human Action Recognition Representations Without Real Humans](https://mlsanity.com/paper/2311.06231/) (2023)
- [ALBAR: Adversarial Learning approach to mitigate Biases in Action Recognition](https://mlsanity.com/paper/2502.00156/) (2025)
- [Scene-aware Learning Network for Radar Object Detection](https://mlsanity.com/paper/2107.01469/) (2021)
- [Evaluating Transformers for Lightweight Action Recognition](https://mlsanity.com/paper/2111.09641/) (2021)
- [Baby Physical Safety Monitoring in Smart Home Using Action Recognition System](https://mlsanity.com/paper/2210.12527/) (2022)
- [Beyond Instance Discrimination: Relation-aware Contrastive Self-supervised Learning](https://mlsanity.com/paper/2211.01796/) (2022)
- [Ctrl-GenAug: Controllable Generative Augmentation for Medical Sequence Classification](https://mlsanity.com/paper/2409.17091/) (2024)
- [Overview of Tencent Multi-modal Ads Video Understanding Challenge](https://mlsanity.com/paper/2109.07951/) (2021)
- [Hate Speech Detection in Limited Data Contexts using Synthetic Data Generation](https://mlsanity.com/paper/2310.02876/) (2023)
- [S3Aug: Segmentation, Sampling, and Shift for Action Recognition](https://mlsanity.com/paper/2310.14556/) (2023)

## Abstract (arXiv)

State-of-the-art video action classifiers often suffer from overfitting. They tend to be biased towards specific objects and scene cues, rather than the foreground action content, leading to sub-optimal generalization performances. Recent data augmentation strategies have been reported to address the overfitting problems in static image classifiers. Despite the effectiveness on the static image classifiers, data augmentation has rarely been studied for videos. For the first time in the field, we systematically analyze the efficacy of various data augmentation strategies on the video classification task. We then propose a powerful augmentation strategy VideoMix. VideoMix creates a new training video by inserting a video cuboid into another video. The ground truth labels are mixed proportionally to the number of voxels from each video. We show that VideoMix lets a model learn beyond the object and scene biases and extract more robust cues for action recognition. VideoMix consistently outperforms other augmentation baselines on Kinetics and the challenging Something-Something-V2 benchmarks. It also improves the weakly-supervised action localization performance on THUMOS'14. VideoMix pretrained models exhibit improved accuracies on the video detection task (AVA).