Researchers have introduced GIFGuard, a novel spatiotemporal watermarking framework designed to combat deepfakes in animated GIFs. This system addresses the limitations of existing forensics tools, which are typically designed for static images. GIFGuard employs a Spatiotemporal Adaptive Residual Encoder (STARE) for robust watermark embedding and a Deep Integrity Restoration Decoder (DIRD) for accurate extraction, even with significant facial manipulation. The project also includes the creation of GIFfaces, a new large-scale dataset for GIF proactive forensics research. AI
影响 Introduces a new method for detecting deepfakes in animated GIFs, potentially improving content authenticity verification.
排序理由 Academic paper introducing a new method and dataset for deepfake detection in GIFs.
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
