- Title
- Anti-aliasing deep image classifiers using novel depth adaptive blurring and activation function
- Creator
- Hossain, Md Tahmid; Teng, Shyh; Lu, Guojun; Rahman, Mohammad Arifur; Sohel, Ferdous
- Date
- 2023
- Type
- Text; Journal article
- Identifier
- http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/196158
- Identifier
- vital:18660
- Identifier
-
https://doi.org/10.1016/j.neucom.2023.03.023
- Identifier
- ISSN:0925-2312 (ISSN)
- Abstract
- Deep convolutional networks are vulnerable to image translation or shift, partly due to common down-sampling layers, e.g., max-pooling and strided convolution. These operations violate the Nyquist sampling rate and cause aliasing. The textbook solution is low-pass filtering (blurring) before down-sampling, which can benefit deep networks as well. Even so, non-linearity units, such as ReLU, often re-introduce the problem, suggesting that blurring alone may not suffice. In this work, first, we analyse deep features with Fourier transform and show that Depth Adaptive Blurring is more effective, as opposed to monotonic blurring. To this end, we propose a novel Depth Adaptive Blur-pool (DAB-pool) module to replace existing down-sampling methods. Second, we introduce a novel activation function – with a built-in low pass filter, as an additional measure, to keep the problem from reappearing. From experiments, we observe generalisation on other forms of transformations and corruptions as well, e.g., rotation, scale, and noise. We evaluate our method under three challenging settings: (1) a variety of image translations; (2) adversarial attacks – both
- Publisher
- Elsevier B.V.
- Relation
- Neurocomputing Vol. 536, no. (2023), p. 164-174
- Rights
- All metadata describing materials held in, or linked to, the repository is freely available under a CC0 licence
- Rights
- 2023 Published by Elsevier B.V.
- Subject
- Anti-aliasing CNN; Convolutional Neural Network (CNN); Corruption; Image noise; Perturbation; Robust CNN; Translation invariant CNN; 40 Engineering; 46 Information and computing sciences; 52 Psychology
- Reviewed
- Funder
- This research was supported by Federation University Research Priority Area scholarship scheme.
- Hits: 999
- Visitors: 921
- Downloads: 0
| Thumbnail | File | Description | Size | Format |
|---|