paper-with-me

홈 › Papers

High-Low Frequency Detectors

2021-01-27 · Distill 2021 1 · Ludwig Schubert, Chelsea Voss, Nick Cammarata, Gabriel Goh, Chris Olah

Some of the neurons in vision models are features that we aren’t particularly surprised to find. Curve detectors, for example, are a pretty natural feature for a vision system to have. In fact, they had already been discovered in the animal visual cortex. It’s easy to imagine how curve detectors are built up from earlier edge detectors, and it’s easy to guess why curve detection might be useful to the rest of the neural network. High-low frequency detectors, on the other hand, seem more surprising. They are not a feature that we would have expected a priori to find. Yet, when systematically characterizing the early layers of InceptionV1, we found a full fifteen neurons of mixed3a that appear to detect a high frequency pattern on one side, and a low frequency pattern on the other. By “high frequency” and “low frequency” here, we mean spatial frequency — just like when we take the Fourier transform of an image. One worry we might have about the circuits approach to studying neural networks is that we might only be able to understand a limited set of highly-intuitive features. High-low frequency detectors demonstrate that it’s possible to understand at least somewhat unintuitive features.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Vocal Bursts Intensity Prediction

Similar Papers 제목 키워드 기반

Exploring Frequency Adversarial Attacks for Face Forgery Detection

2022-03-29 · CVPR 2022 1 · Shuai Jia, Chao Ma, Taiping Yao, Bangjie Yin 외

Various facial manipulation techniques have drawn serious public concerns in morality, security, and privacy. Although existing face forgery classifiers achieve promising performance on detecting fake images, these metho…

Adversarial AttackMeta-Learning

Dual Data Alignment Makes AI-Generated Image Detector Easier Generalizable

2025-05-20 · Ruoxin Chen, Junwei Xi, Zhiyuan Yan, Ke-Yue Zhang 외

Existing detectors are often trained on biased datasets, leading to the possibility of overfitting on non-causal image attributes that are spuriously correlated with real/synthetic labels. While these biased features enh…

Beyond Spatial Frequency: Pixel-wise Temporal Frequency-based Deepfake Video Detection

2025-07-03 · TaeHoon Kim, Jongwook Choi, Yonghyun Jeong, Haeun Noh 외

We introduce a deepfake video detection approach that exploits pixel-wise temporal inconsistencies, which traditional spatial frequency-based detectors often overlook. Traditional detectors represent temporal information…

Face Swapping

Vulnerabilities in AI-generated Image Detection: The Challenge of Adversarial Attacks

2024-07-30 · Yunfeng Diao, Naixin Zhai, Changtao Miao, Zitong Yu 외

Recent advancements in image synthesis, particularly with the advent of GAN and Diffusion models, have amplified public concerns regarding the dissemination of disinformation. To address such concerns, numerous AI-genera…

Adversarial AttackAdversarial RobustnessImage Generation

Enhancing Frequency Forgery Clues for Diffusion-Generated Image Detection

2025-11-01 · Daichi Zhang, Tong Zhang, Shiming Ge, Sabine Süsstrunk arxiv

Diffusion models have achieved remarkable success in image synthesis, but the generated high-quality images raise concerns about potential malicious use. Existing detectors often struggle to capture discriminative clues …