# Dynamic Network selection for the Object Detection task: why it matters and what we (didn't) achieve

arXiv: https://arxiv.org/abs/2105.13279 · HTML: https://mlsanity.com/paper/2105.13279/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 25 references that are in the map.
- First posted on arXiv 27 May 2021.

## Builds on (most cited first, 8 of 8)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Rich feature hierarchies for accurate object detection and semantic segmentation](https://mlsanity.com/paper/1311.2524/) (2013)
- [MnasNet: Platform-Aware Neural Architecture Search for Mobile](https://mlsanity.com/paper/1807.11626/) (2018)
- [Deep Speech 2: End-to-End Speech Recognition in English and Mandarin](https://mlsanity.com/paper/1512.02595/) (2015)
- [Speed/accuracy trade-offs for modern convolutional object detectors](https://mlsanity.com/paper/1611.10012/) (2016)
- [Flexible Deep Neural Network Processing](https://mlsanity.com/paper/1801.07353/) (2018)
- [An Efficient Monte Carlo-based Probabilistic Time-Dependent Routing Calculation Targeting a Server-Side Car Navigation System](https://mlsanity.com/paper/1901.06210/) (2019)

## Abstract (arXiv)

In this paper, we want to show the potential benefit of a dynamic auto-tuning approach for the inference process in the Deep Neural Network (DNN) context, tackling the object detection challenge. We benchmarked different neural networks to find the optimal detector for the well-known COCO 17 database, and we demonstrate that even if we only consider the quality of the prediction there is not a single optimal network. This is even more evident if we also consider the time to solution as a metric to evaluate, and then select, the most suitable network. This opens to the possibility for an adaptive methodology to switch among different object detection networks according to run-time requirements (e.g. maximum quality subject to a time-to-solution constraint). Moreover, we demonstrated by developing an ad hoc oracle, that an additional proactive methodology could provide even greater benefits, allowing us to select the best network among the available ones given some characteristics of the processed image. To exploit this method, we need to identify some image features that can be used to steer the decision on the most promising network. Despite the optimization opportunity that has been identified, we were not able to identify a predictor function that validates this attempt neither adopting classical image features nor by using a DNN classifier.