Author name cluster

Simon Woo

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

1 author row

NeurIPS Conference 2025 Conference Paper

RUAGO: Effective and Practical Retain-Free Unlearning via Adversarial Attack and OOD Generator

Sangyong Lee
Sangjun Chung
Simon Woo

With increasing regulations on private data usage in AI systems, machine unlearning has emerged as a critical solution for selectively removing sensitive information from trained models while preserving their overall utility. While many existing unlearning methods rely on the retain data to mitigate the performance decline caused by forgetting, such data may not always be available ( retain-free ) in real-world scenarios. To address this challenge posed by retain-free unlearning, we introduce RUAGO, utilizing adversarial soft labels to mitigate over-unlearning and a generative model pretrained on out-of-distribution (OOD) data to effectively distill the original model’s knowledge. We introduce a progressive sampling strategy to incrementally increase synthetic data complexity, coupled with an inversion-based alignment step that ensures the synthetic data closely matches the original training distribution. Our extensive experiments on multiple benchmark datasets and architectures demonstrate that our approach consistently outperforms existing retain-free methods and achieves comparable or superior performance relative to retain-based approaches, demonstrating its effectiveness and practicality in real-world, data-constrained environments.

PDF Details

NeurIPS Conference 2025 Conference Paper

Through the Lens: Benchmarking Deepfake Detectors Against Moiré-Induced Distortions

Razaib Tariq
Minji Heo
Shahroz Tariq
Simon Woo

Deepfake detection remains a pressing challenge, particularly in real-world settings where smartphone-captured media from digital screens often introduces Moiré artifacts that can distort detection outcomes. This study systematically evaluates state-of-the-art (SOTA) deepfake detectors on Moiré-affected videos—an issue that has received little attention. We collected a dataset of 12, 832 videos, spanning 35. 64 hours, from Celeb-DF, DFD, DFDC, UADFV, and FF++ datasets, capturing footage under diverse real-world conditions, including varying screens, smartphones, lighting setups, and camera angles. To further examine the influence of Moiré patterns on deepfake detection, we conducted additional experiments using our DeepMoiréFake, referred to as (DMF) dataset, and two synthetic Moiré generation techniques. Across 15 top-performing detectors, our results show that Moiré artifacts degrade performance by as much as 25. 4\%, while synthetically generated Moiré patterns lead to a 21. 4\% drop in accuracy. Surprisingly, demoiréing methods, intended as a mitigation approach, instead worsened the problem, reducing accuracy by up to 16\%. These findings underscore the urgent need for detection models that can robustly handle Moiré distortions alongside other real-world challenges, such as compression, sharpening, and blurring. By introducing the DMF dataset, we aim to drive future research toward closing the gap between controlled experiments and practical deepfake detection.

PDF Details

AAAI Conference 2022 Conference Paper

ADD: Frequency Attention and Multi-View Based Knowledge Distillation to Detect Low-Quality Compressed Deepfake Images

Le Minh Binh
Simon Woo

Despite significant advancements of deep learning-based forgery detectors for distinguishing manipulated deepfake images, most detection approaches suffer from moderate to significant performance degradation with low-quality compressed deepfake images. Because of the limited information in low-quality images, detecting low-quality deepfake remains an important challenge. In this work, we apply frequency domain learning and optimal transport theory in knowledge distillation (KD) to specifically improve the detection of low-quality compressed deepfake images. We explore transfer learning capability in KD to enable a student network to learn discriminative features from low-quality images effectively. In particular, we propose the Attention-based Deepfake detection Distiller (ADD), which consists of two novel distillations: 1) frequency attention distillation that effectively retrieves the removed high-frequency components in the student network, and 2) multi-view attention distillation that creates multiple attention vectors by slicing the teacher’s and student’s tensors under different views to transfer the teacher tensor’s distribution to the student more efficiently. Our extensive experimental results demonstrate that our approach outperforms state-of-the-art baselines in detecting low-quality compressed deepfake images.

PDF Details

NeurIPS Conference 2021 Conference Paper

FakeAVCeleb: A Novel Audio-Video Multimodal Deepfake Dataset

Hasam Khalid
Shahroz Tariq
Minha Kim
Simon Woo

While the significant advancements have made in the generation of deepfakes using deep learning technologies, its misuse is a well-known issue now. Deepfakes can cause severe security and privacy issues as they can be used to impersonate a person's identity in a video by replacing his/her face with another person's face. Recently, a new problem of generating synthesized human voice of a person is emerging, where AI-based deep learning models can synthesize any person's voice requiring just a few seconds of audio. With the emerging threat of impersonation attacks using deepfake audios and videos, a new generation of deepfake detectors is needed to focus on both video and audio collectively. To develop a competent deepfake detector, a large amount of high-quality data is typically required to capture real-world (or practical) scenarios. Existing deepfake datasets either contain deepfake videos or audios, which are racially biased as well. As a result, it is critical to develop a high-quality video and audio deepfake dataset that can be used to detect both audio and video deepfakes simultaneously. To fill this gap, we propose a novel Audio-Video Deepfake dataset, FakeAVCeleb, which contains not only deepfake videos but also respective synthesized lip-synced fake audios. We generate this dataset using the current most popular deepfake generation methods. We selected real YouTube videos of celebrities with four ethnic backgrounds to develop a more realistic multimodal dataset that addresses racial bias, and further help develop multimodal deepfake detectors. We performed several experiments using state-of-the-art detection methods to evaluate our deepfake dataset and demonstrate the challenges and usefulness of our multimodal Audio-Video deepfake dataset.

PDF Details

NeurIPS Conference 2021 Conference Paper

VFP290K: A Large-Scale Benchmark Dataset for Vision-based Fallen Person Detection

Jaeju An
Jeongho Kim
Hanbeen Lee
Jinbeom Kim
Junhyung Kang
Saebyeol Shin
Minha Kim
Donghee Hong

Detection of fallen persons due to, for example, health problems, violence, or accidents, is a critical challenge. Accordingly, detection of these anomalous events is of paramount importance for a number of applications, including but not limited to CCTV surveillance, security, and health care. Given that many detection systems rely on a comprehensive dataset comprising fallen person images collected under diverse environments and in various situations is crucial. However, existing datasets are limited to only specific environmental conditions and lack diversity. To address the above challenges and help researchers develop more robust detection systems, we create a novel, large-scale dataset for the detection of fallen persons composed of fallen person images collected in various real-world scenarios, with the support of the South Korean government. Our Vision-based Fallen Person (VFP290K) dataset consists of 294, 713 frames of fallen persons extracted from 178 videos, including 131 scenes in 49 locations. We empirically demonstrate the effectiveness of the features through extensive experiments analyzing the performance shift based on object detection models. In addition, we evaluate our VFP290K dataset with properly divided versions of our dataset by measuring the performance of fallen person detecting systems. We ranked first in the first round of the anomalous behavior recognition track of AI Grand Challenge 2020, South Korea, using our VFP290K dataset, which can be found here. Our achievement implies the usefulness of our dataset for research on fallen person detection, which can further extend to other applications, such as intelligent CCTV or monitoring systems. The data and more up-to-date information have been provided at our VFP290K site.

PDF Details