WAVIE improves deepfake detection by combining spatial and frequency cues
Lightweight Generalized DeepFake Face Detection with WAVIE: Wavelet Augmented Vision Intermediate Embeddings
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Detecting fake videos called deepfakes is hard because systems often fail when they see new types of fakes they weren’t trained on. The authors present WAVIE, a new method that looks at video images using both their visual details and hidden frequency patterns. By using a special wavelet transformation on features extracted from a powerful image model, WAVIE better spots fake videos it hasn’t seen before. It works well on several challenging test sets, showing promise for more reliable deepfake detection.
What this means in practice
- •For digital forensics teams: Improve detection of manipulated face videos across varied and unseen deepfake methods in forensic investigations.
- •For content moderation platforms: Enhance automated screening of user-generated videos for deepfake manipulations not included in training data.
Authors
Arya Pulkit, Aditya Ruhela, Akarshan Kapoor, Arnav Bhavsar
Abstract
Deepfake detection systems often exhibit significant performance degradation when deployed on unseen manipulation methods, limiting their reliability in real-world multimedia environments. This lack of generalization poses critical challenges for misinformation mitigation, digital forensics, and human-centric AI systems. Existing detectors perform well on the forgery methods they are trained on, but their accuracy drops sharply on unseen pipelines. To bridge this generalization gap, we propose WAVIE (Wavelet Augmented Vision Intermediate Embeddings), an end-to-end architecture that combines complementary spatial and frequency cues on top of a frozen CLIP backbone. WAVIE projects intermediate transformer embeddings through a lightweight learnable module, applies a three-level Daubechies-6 (db6) discrete wavelet transform (DWT), refines the low-frequency branch while preserving the high-frequency branch, reconstructs the feature via inverse DWT, and performs classification. Trained only on FaceForensics++, WAVIE achieves AUROC = 0.852 on Celeb-DF-v1, 0.852 on Celeb-DF-v2 and 0.831 on WildDeepFake (WDF) at the frame level, outperforming several state-of-the-art generalization baselines. Extensive ablation studies confirm the importance of both the wavelet module and the intermediate-feature aggregation for cross-dataset performance, highlighting the necessity of jointly leveraging spatial and frequency domains. These results position WAVIE as a strong baseline for deepfake detection in the wild.