Investigating and Explaining the Frequency Bias in Image Classification
- URL: http://arxiv.org/abs/2205.03154v1
- Date: Fri, 6 May 2022 11:45:43 GMT
- Title: Investigating and Explaining the Frequency Bias in Image Classification
- Authors: ZhiYu Lin, YiFei Gao, JiTao Sang
- Abstract summary: CNNs exhibit many behaviors different from humans, one of which is the capability of employing high-frequency components.
This paper discusses the frequency bias phenomenon in image classification tasks.
- Score: 11.078920943157845
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: CNNs exhibit many behaviors different from humans, one of which is the
capability of employing high-frequency components. This paper discusses the
frequency bias phenomenon in image classification tasks: the high-frequency
components are actually much less exploited than the low- and mid-frequency
components. We first investigate the frequency bias phenomenon by presenting
two observations on feature discrimination and learning priority. Furthermore,
we hypothesize that (i) the spectral density, (ii) class consistency directly
affect the frequency bias. Specifically, our investigations verify that the
spectral density of datasets mainly affects the learning priority, while the
class consistency mainly affects the feature discrimination.
Related papers
- Tuning Frequency Bias of State Space Models [48.60241978021799]
State space models (SSMs) leverage linear, time-invariant (LTI) systems to learn sequences with long-range dependencies.
We find that SSMs exhibit an implicit bias toward capturing low-frequency components more effectively than high-frequency ones.
arXiv Detail & Related papers (2024-10-02T21:04:22Z) - Mitigating Low-Frequency Bias: Feature Recalibration and Frequency Attention Regularization for Adversarial Robustness [23.77988226456179]
This paper proposes a novel module called High-Frequency Feature Disentanglement and Recalibration (HFDR)
HFDR separates features into high-frequency and low-frequency components and recalibrates the high-frequency feature to capture latent useful semantics.
Extensive experiments showcase the immense potential and superiority of our approach in resisting various white-box attacks, transfer attacks, and showcasing strong generalization capabilities.
arXiv Detail & Related papers (2024-07-04T15:46:01Z) - FS-BAND: A Frequency-Sensitive Banding Detector [55.59101150019851]
Banding artifact, as known as staircase-like contour, is a common quality annoyance that happens in compression, transmission, etc.
We propose a no-reference banding detection model to capture and evaluate banding artifacts, called the Frequency-Sensitive BANding Detector (FS-BAND)
Experimental results show that the proposed FS-BAND method outperforms state-of-the-art image quality assessment (IQA) approaches with higher accuracy in banding classification task.
arXiv Detail & Related papers (2023-11-30T03:20:42Z) - What do neural networks learn in image classification? A frequency
shortcut perspective [3.9858496473361402]
This study empirically investigates the learning dynamics of frequency shortcuts in neural networks (NNs)
We show that NNs tend to find simple solutions for classification, and what they learn first during training depends on the most distinctive frequency characteristics.
We propose a metric to measure class-wise frequency characteristics and a method to identify frequency shortcuts.
arXiv Detail & Related papers (2023-07-19T08:34:25Z) - On the Frequency Bias of Generative Models [61.60834513380388]
We analyze proposed measures against high-frequency artifacts in state-of-the-art GAN training.
We find that none of the existing approaches can fully resolve spectral artifacts yet.
Our results suggest that there is great potential in improving the discriminator.
arXiv Detail & Related papers (2021-11-03T18:12:11Z) - A Frequency Perspective of Adversarial Robustness [72.48178241090149]
We present a frequency-based understanding of adversarial examples, supported by theoretical and empirical findings.
Our analysis shows that adversarial examples are neither in high-frequency nor in low-frequency components, but are simply dataset dependent.
We propose a frequency-based explanation for the commonly observed accuracy vs. robustness trade-off.
arXiv Detail & Related papers (2021-10-26T19:12:34Z) - Spectral Bias in Practice: The Role of Function Frequency in
Generalization [10.7218588164913]
We propose methodologies for measuring spectral bias in modern image classification networks.
We find that networks that generalize well strike a balance between having enough complexity to fit the data while being simple enough to avoid overfitting.
Our work enables measuring and ultimately controlling the spectral behavior of neural networks used for image classification.
arXiv Detail & Related papers (2021-10-06T00:16:10Z) - Impact of Spatial Frequency Based Constraints on Adversarial Robustness [0.49478969093606673]
Adversarial examples mainly exploit changes to input pixels to which humans are not sensitive, and arise from the fact that models make decisions based on uninterpretable features.
In this paper, we investigate the robustness to adversarial perturbations of models enforced during training to leverage information corresponding to different spatial frequency ranges.
arXiv Detail & Related papers (2021-04-26T16:12:04Z) - Generalizing Face Forgery Detection with High-frequency Features [63.33397573649408]
Current CNN-based detectors tend to overfit to method-specific color textures and thus fail to generalize.
We propose to utilize the high-frequency noises for face forgery detection.
The first is the multi-scale high-frequency feature extraction module that extracts high-frequency noises at multiple scales.
The second is the residual-guided spatial attention module that guides the low-level RGB feature extractor to concentrate more on forgery traces from a new perspective.
arXiv Detail & Related papers (2021-03-23T08:19:21Z) - WaveTransform: Crafting Adversarial Examples via Input Decomposition [69.01794414018603]
We introduce WaveTransform', that creates adversarial noise corresponding to low-frequency and high-frequency subbands, separately (or in combination)
Experiments show that the proposed attack is effective against the defense algorithm and is also transferable across CNNs.
arXiv Detail & Related papers (2020-10-29T17:16:59Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.