[1611.10012] Speed/accuracy trade-offs for modern convolutional object detectorscontact arXivarXiv Twitter

The goal of this paper is to serve as a guide for selecting a detection architecture that achieves the right speed/memory/accuracy balance for a given application and platform. To this end, we investigate various ways to trade accuracy for speed and memory usage in modern convolutional object detection systems. A number of successful systems have been proposed in recent years, but apples-to-apples comparisons are difficult due to different base feature extractors (e.g., VGG, Residual Networks), different default image resolutions, as well as different hardware and software platforms. We present a unified implementation of the Faster R-CNN [Ren et al., 2015], R-FCN [Dai et al., 2016] and SSD [Liu et al., 2015] systems, which we view as "meta-architectures" and trace out the speed/accuracy trade-off curve created by using alternative feature extractors and varying other critical parameters such as image size within each of these meta-architectures. On one extreme end of this spectrum whe

1 mentions: @izariuo440
Date: 2020/02/15 05:21

Bookmark Comments

Related Entries

Read more Hyperparameter optimization for Neural Networks — NeuPy
22 users, 1 mentions 2019/02/21 08:17
Read more [1905.07299] Spectral Metric for Dataset Complexity Assessment
2 users, 1 mentions 2019/10/02 03:48
Read more Kaiming He - FAIR
5 users, 1 mentions 2019/10/04 02:18
Read more [1308.0850] Generating Sequences With Recurrent Neural Networks
11 users, 1 mentions 2019/10/23 11:18
Read more PyMC3 Documentation — PyMC3 3.8 documentation
4 users, 1 mentions 2020/02/12 08:21