Robust Deepfake Detection: Mitigating Spatial Attention Drift via Calibrated Complementary Ensembles

• URL: http://arxiv.org/abs/2604.25889v1
• Date: Tue, 28 Apr 2026 17:32:48 GMT
• Title: Robust Deepfake Detection: Mitigating Spatial Attention Drift via Calibrated Complementary Ensembles
• Abstract summary: Current deepfake detection models suffer from spatial attention drift under real-world compound degradations.<n>We propose a foundation-driven forensic framework that integrates an extreme compound degradation engine with a structurally constrained, multi-stream architecture.<n>Our approach yields highly stable zero-shot generalization, achieving Fourth Place in the NTIRE 2026 Robust Deepfake Detection Challenge at CVPR.
• Score: 8.588817718552418
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Current deepfake detection models achieve state-of-the-art performance on pristine academic datasets but suffer severe spatial attention drift under real-world compound degradations, such as blurring and severe lossy compression. To address this vulnerability, we propose a foundation-driven forensic framework that integrates an extreme compound degradation engine with a structurally constrained, multi-stream architecture. During training, our degradation pipeline systematically destroys high-frequency artifacts, optimizing the DINOv2-Giant backbone to extract invariant geometric and semantic priors. We then process images through three specialized pathways: a Global Texture stream, a Localized Facial stream, and a Hybrid Semantic Fusion stream incorporating CLIP. Through analyzing spatial attribution via Score-CAM and feature stability using Cosine Similarity, we quantitatively demonstrate that these streams extract non-redundant, complementary feature representations and stabilize attention entropy. By aggregating these predictions via a calibrated, discretized voting mechanism, our ensemble successfully suppresses background attention drift while acting as a robust geometric anchor. Our approach yields highly stable zero-shot generalization, achieving Fourth Place in the NTIRE 2026 Robust Deepfake Detection Challenge at CVPR. Code is available at https://github.com/khoalephanminh/ntire26-deepfake-challenge.

Related papers

• Need We Teach Foundation Models What is a Generative Image? Gradient-Free Generative Artifact Detection via Analytic Spectral Adaptation [3.881211374324378]
  Adapting foundation models to detect generative artifacts via gradient-based updates compromises their intrinsic representations.<n>We propose a gradient-free methodology reframing detection from binary classification to an out-of-distribution (OOD) anomaly measurement problem.
  arXiv  Detail & Related papers  (2026-06-03T13:04:23Z)
• Hierarchical Consistency Learning for Test-time Adaptation in Camouflage Perception [50.278200968044665]
  Camouflaged object detection (COD) aims to localize targets that exhibit minimal perceptual differences from backgrounds through physical attributes.<n>Existing methods, constrained by the static train-then-freeze paradigm, suffer from domain rigidity and dependency annotation.<n>We propose the hierarchical consistency learning framework, which integrates test-time adaptation for dynamic representation recalibration.
  arXiv  Detail & Related papers  (2026-05-25T09:57:46Z)
• DOC-GS: Dual-Domain Observation and Calibration for Reliable Sparse-View Gaussian Splatting [80.43237927269575]
  In this paper, we revisit sparse-view 3DGS reconstruction from a new perspective.<n>We identify the core challenge as the unobservability of Gaussian primitive reliability.<n>Motivated by this observation, we propose a unified Dual-domain Observation and geometrics in rendered images framework.
  arXiv  Detail & Related papers  (2026-04-08T07:01:24Z)
• ForensicFlow: A Tri-Modal Adaptive Network for Robust Deepfake Detection [0.0]
  We introduce ForensicFlow, a tri-modal forensic framework that fuses RGB, texture, and frequency evidence for video Deepfake detection.<n>Trained on Celeb-DF (v2) with Focal Loss, ForensicFlow achieves AUC 0.9752, F1-Score 0.9408, and accuracy 0.9208, outperforming single-stream baselines.
  arXiv  Detail & Related papers  (2025-11-18T14:56:34Z)
• Flow-Matching Guided Deep Unfolding for Hyperspectral Image Reconstruction [53.26903617819014]
  Flow-Matching-guided Unfolding network (FMU) is first to integrate flow matching into HSI reconstruction.<n>To further strengthen the learned dynamics, we introduce a mean velocity loss.<n>Experiments on both simulated and real datasets show that FMU significantly outperforms existing approaches in reconstruction quality.
  arXiv  Detail & Related papers  (2025-10-02T11:32:00Z)
• Deep Equilibrium Convolutional Sparse Coding for Hyperspectral Image Denoising [16.405355853358202]
  Hyperspectral images (HSIs) play a crucial role in remote sensing but are often degraded by complex noise patterns.<n> Ensuring the physical property of the denoised HSIs is vital for robust HSI denoising, giving the rise of deep unfolding-based methods.<n>We propose a Deep Equilibrium Convolutional Sparse Coding (DECSC) framework that unifies local spatial-spectral correlations, nonlocal spatial self-similarities, and global spatial consistency.
  arXiv  Detail & Related papers  (2025-08-21T13:35:11Z)
• JointSplat: Probabilistic Joint Flow-Depth Optimization for Sparse-View Gaussian Splatting [10.690965024885358]
  Reconstructing 3D scenes from sparse viewpoints is a long-standing challenge with wide applications.<n>Recent advances in feed-forward 3D Gaussian sparse-view reconstruction methods provide an efficient solution for real-time novel view synthesis.<n>We propose JointSplat, a unified framework that leverages the complementarity between optical flow and depth.
  arXiv  Detail & Related papers  (2025-06-04T12:04:40Z)
• GC-ConsFlow: Leveraging Optical Flow Residuals and Global Context for Robust Deepfake Detection [23.05634856249282]
  Deepfake technology has enabled the generation of highly realistic manipulated videos, posing severe social and ethical challenges.<n>Existing Deepfake detection methods primarily focused on either spatial or temporal inconsistencies, neglecting the interplay between the two.<n>We propose the global context flow (GC-ConsFlow), a novel dual-stream framework that effectively integrates spatial and temporal features for robust Deepfake detection.
  arXiv  Detail & Related papers  (2025-01-23T07:43:56Z)
• HazyDet: Open-Source Benchmark for Drone-View Object Detection with Depth-Cues in Hazy Scenes [54.24350833692194]
  HazyDet is the first, large-scale benchmark specifically designed for drone-view object detection in hazy conditions.<n>We propose the Depth-Conditioned Detector (DeCoDet) to address the severe visual degradation induced by haze.<n>HazyDet provides a challenging and realistic testbed for advancing detection algorithms.
  arXiv  Detail & Related papers  (2024-09-30T00:11:40Z)
• On Robust Cross-View Consistency in Self-Supervised Monocular Depth Estimation [56.97699793236174]
  We study two kinds of robust cross-view consistency in this paper. We exploit the temporal coherence in both depth feature space and 3D voxel space for self-supervised monocular depth estimation. Experimental results on several outdoor benchmarks show that our method outperforms current state-of-the-art techniques.
  arXiv  Detail & Related papers  (2022-09-19T03:46:13Z)
• The KFIoU Loss for Rotated Object Detection [115.334070064346]
  In this paper, we argue that one effective alternative is to devise an approximate loss who can achieve trend-level alignment with SkewIoU loss. Specifically, we model the objects as Gaussian distribution and adopt Kalman filter to inherently mimic the mechanism of SkewIoU. The resulting new loss called KFIoU is easier to implement and works better compared with exact SkewIoU.
  arXiv  Detail & Related papers  (2022-01-29T10:54:57Z)

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.