# THIA: Accelerating Video Analytics using Early Inference and Fine-Grained Query Planning

arXiv: https://arxiv.org/abs/2102.08481 · HTML: https://mlsanity.com/paper/2102.08481/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Feb 2021.

## Built on by (most cited first, 8 of 8)

- [Edge Video Analytics: A Survey on Applications, Systems and Enabling Techniques](https://mlsanity.com/paper/2211.15751/) (2022)
- [Extract-Transform-Load for Video Streams](https://mlsanity.com/paper/2310.04830/) (2023)
- [CoVA: Exploiting Compressed-Domain Analysis to Accelerate Video Analytics](https://mlsanity.com/paper/2207.00588/) (2022)
- [Saturn: An Optimized Data System for Large Model Deep Learning Workloads](https://mlsanity.com/paper/2309.01226/) (2023)
- [VQPy: An Object-Oriented Approach to Modern Video Analytics](https://mlsanity.com/paper/2311.01623/) (2023)
- [A Survey of Performance Optimization in Neural Network-Based Video Analytics Systems](https://mlsanity.com/paper/2105.14195/) (2021)
- [Automatic and Efficient Customization of Neural Networks for ML Applications](https://mlsanity.com/paper/2310.04685/) (2023)
- [Responsive DNN Adaptation for Video Analytics against Environment Shift via Hierarchical Mobile-Cloud Collaborations](https://mlsanity.com/paper/2505.00745/) (2025)

## Abstract (arXiv)

To efficiently process visual data at scale, researchers have proposed two techniques for lowering the computational overhead associated with the underlying deep learning models. The first approach consists of leveraging a specialized, lightweight model to directly answer the query. The second approach focuses on filtering irrelevant frames using a lightweight model and processing the filtered frames using a heavyweight model. These techniques suffer from two limitations. With the first approach, the specialized model is unable to provide accurate results for hard-to-detect events. With the second approach, the system is unable to accelerate queries focusing on frequently occurring events as the filter is unable to eliminate a significant fraction of frames in the video. In this paper, we present THIA, a video analytics system for tackling these limitations. The design of THIA is centered around three techniques. First, instead of using a cascade of models, it uses a single object detection model with multiple exit points for short-circuiting the inference. This early inference technique allows it to support a range of throughput-accuracy tradeoffs. Second, it adopts a fine-grained approach to planning and processes different chunks of the video using different exit points to meet the user's requirements. Lastly, it uses a lightweight technique for directly estimating the exit point for a chunk to lower the optimization time. We empirically show that these techniques enable THIA to outperform two state-of-the-art video analytics systems by up to 6.5X, while providing accurate results even on queries focusing on hard-to-detect events.