Adaptations to CSP including compressing width prediction achieve 9.3% MR on CityPersons reasonable set, showing anchor-free one-stage detectors can reach high accuracy.
Differentiable Learning-to-Normalize via Switchable Normalization
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
We address a learning-to-normalize problem by proposing Switchable Normalization (SN), which learns to select different normalizers for different normalization layers of a deep neural network. SN employs three distinct scopes to compute statistics (means and variances) including a channel, a layer, and a minibatch. SN switches between them by learning their importance weights in an end-to-end manner. It has several good properties. First, it adapts to various network architectures and tasks (see Fig.1). Second, it is robust to a wide range of batch sizes, maintaining high performance even when small minibatch is presented (e.g. 2 images/GPU). Third, SN does not have sensitive hyper-parameter, unlike group normalization that searches the number of groups as a hyper-parameter. Without bells and whistles, SN outperforms its counterparts on various challenging benchmarks, such as ImageNet, COCO, CityScapes, ADE20K, and Kinetics. Analyses of SN are also presented. We hope SN will help ease the usage and understand the normalization techniques in deep learning. The code of SN has been made available in https://github.com/switchablenorms/.
fields
cs.CV 1years
2020 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Adapted Center and Scale Prediction: More Stable and More Accurate
Adaptations to CSP including compressing width prediction achieve 9.3% MR on CityPersons reasonable set, showing anchor-free one-stage detectors can reach high accuracy.