# Breaking the accuracy-resource dilemma: a lightweight adaptive video inference enhancement

arXiv: https://arxiv.org/abs/2601.14568 · HTML: https://mlsanity.com/paper/2601.14568/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 26 references that are in the map.
- First posted on arXiv 21 Jan 2026.

## Builds on (most cited first, 14 of 14)

- [Detection and Tracking Meet Drones Challenge](https://mlsanity.com/paper/2001.06303/) (2020)
- [A Survey on Mixture of Experts in Large Language Models](https://mlsanity.com/paper/2407.06204/) (2024)
- [Uni-MoE: Scaling Unified Multimodal LLMs with Mixture of Experts](https://mlsanity.com/paper/2405.11273/) (2024)
- [UA-DETRAC: A New Benchmark and Protocol for Multi-Object Detection and Tracking](https://mlsanity.com/paper/1511.04136/) (2015)
- [Lenna: Language Enhanced Reasoning Detection Assistant](https://mlsanity.com/paper/2312.02433/) (2023)
- [An Efficient Hardware Accelerator for Structured Sparse Convolutional Neural Networks on FPGAs](https://mlsanity.com/paper/2001.01955/) (2020)
- [Towards Self-Adaptive Machine Learning-Enabled Systems Through QoS-Aware Model Switching](https://mlsanity.com/paper/2308.09960/) (2023)
- [WHC: Weighted Hybrid Criterion for Filter Pruning on Convolutional Neural Networks](https://mlsanity.com/paper/2302.08185/) (2023)
- [SWITCH: An Exemplar for Evaluating Self-Adaptive ML-Enabled Systems](https://mlsanity.com/paper/2402.06351/) (2024)
- [Mixture of Experts Guided by Gaussian Splatters Matters: A new Approach to Weakly-Supervised Video Anomaly Detection](https://mlsanity.com/paper/2508.06318/) (2025)
- [POCKET: Pruning Random Convolution Kernels for Time Series Classification from a Feature Selection Perspective](https://mlsanity.com/paper/2309.08499/) (2023)
- [EdgeMLBalancer: A Self-Adaptive Approach for Dynamic Model Switching on Resource-Constrained Edge Devices](https://mlsanity.com/paper/2502.06493/) (2025)
- [InternVQA: Advancing Compressed Video Quality Assessment with Distilling Large Foundation Model](https://mlsanity.com/paper/2502.19026/) (2025)
- [ZS-VCOS: Zero-Shot Video Camouflaged Object Segmentation By Optical Flow and Open Vocabulary Object Detection](https://mlsanity.com/paper/2505.01431/) (2025)

## Abstract (arXiv)

Existing video inference (VI) enhancement methods typically aim to improve performance by scaling up model sizes and employing sophisticated network architectures. While these approaches demonstrated state-of-the-art performance, they often overlooked the trade-off of resource efficiency and inference effectiveness, leading to inefficient resource utilization and suboptimal inference performance. To address this problem, a fuzzy controller (FC-r) is developed based on key system parameters and inference-related metrics. Guided by the FC-r, a VI enhancement framework is proposed, where the spatiotemporal correlation of targets across adjacent video frames is leveraged. Given the real-time resource conditions of the target device, the framework can dynamically switch between models of varying scales during VI. Experimental results demonstrate that the proposed method effectively achieves a balance between resource utilization and inference performance.