# Fast Video Classification via Adaptive Cascading of Deep Models

arXiv: https://arxiv.org/abs/1611.06453 · HTML: https://mlsanity.com/paper/1611.06453/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Nov 2016.

## Built on by (most cited first, 17 of 17)

- [Arithmetic-Intensity-Guided Fault Tolerance for Neural Network Inference on GPUs](https://mlsanity.com/paper/2104.09455/) (2021)
- [Deep Neural Network-based Enhancement for Image and Video Streaming Systems: A Survey and Future Directions](https://mlsanity.com/paper/2106.03727/) (2021)
- [Accelerating Deep Learning Inference via Learned Caches](https://mlsanity.com/paper/2101.07344/) (2021)
- [Context-Aware Streaming Perception in Dynamic Environments](https://mlsanity.com/paper/2208.07479/) (2022)
- [Edge-Based Video Analytics: A Survey](https://mlsanity.com/paper/2303.14329/) (2023)
- [AccMPEG: Optimizing Video Encoding for Video Analytics](https://mlsanity.com/paper/2204.12534/) (2022)
- [Turbo: Opportunistic Enhancement for Edge Video Analytics](https://mlsanity.com/paper/2207.00172/) (2022)
- [CoVA: Exploiting Compressed-Domain Analysis to Accelerate Video Analytics](https://mlsanity.com/paper/2207.00588/) (2022)
- [CascadeServe: Unlocking Model Cascades for Inference Serving](https://mlsanity.com/paper/2406.14424/) (2024)
- [Towards Automatic Model Specialization for Edge Video Analytics](https://mlsanity.com/paper/2104.06826/) (2021)
- [WildFit: Autonomous In-situ Model Adaptation for Resource-Constrained IoT Systems](https://mlsanity.com/paper/2409.07796/) (2024)
- [Pool of Experts: Realtime Querying Specialized Knowledge in Massive Neural Networks](https://mlsanity.com/paper/2107.01354/) (2021)
- [Characterizing and Taming Resolution in Convolutional Neural Networks](https://mlsanity.com/paper/2110.14819/) (2021)
- [Automatic and Efficient Customization of Neural Networks for ML Applications](https://mlsanity.com/paper/2310.04685/) (2023)
- [Efficient IoT Inference via Context-Awareness](https://mlsanity.com/paper/2310.19112/) (2023)
- [Legilimens: Performant Video Analytics on the System-on-Chip Edge](https://mlsanity.com/paper/2504.21136/) (2025)
- [Responsive DNN Adaptation for Video Analytics against Environment Shift via Hierarchical Mobile-Cloud Collaborations](https://mlsanity.com/paper/2505.00745/) (2025)

## Abstract (arXiv)

Recent advances have enabled "oracle" classifiers that can classify across many classes and input distributions with high accuracy without retraining. However, these classifiers are relatively heavyweight, so that applying them to classify video is costly. We show that day-to-day video exhibits highly skewed class distributions over the short term, and that these distributions can be classified by much simpler models. We formulate the problem of detecting the short-term skews online and exploiting models based on it as a new sequential decision making problem dubbed the Online Bandit Problem, and present a new algorithm to solve it. When applied to recognizing faces in TV shows and movies, we realize end-to-end classification speedups of 2.4-7.8x/2.6-11.2x (on GPU/CPU) relative to a state-of-the-art convolutional neural network, at competitive accuracy.