Summary of Faster Than Lies: Real-time Deepfake Detection Using Binary Neural Networks, by Lanzino Romeo et al.
Faster Than Lies: Real-time Deepfake Detection using Binary Neural Networks
by Lanzino Romeo, Fontana Federico, Diko Anxhelo, Marini Marco Raoul, Cinque Luigi
First submitted to arxiv on: 7 Jun 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper introduces a novel approach to deepfake detection on images using Binary Neural Networks (BNNs) that prioritizes fast inference with minimal accuracy loss. Unlike previous methods focused on large and complex models, this BNN-based method leverages FFT and LBP channel features to uncover manipulation traces in frequency and texture domains. The proposed approach demonstrates state-of-the-art performance on COCOFake, DFFD, and CIFAKE datasets, achieving a significant efficiency gain of up to a 20x reduction in FLOPs during inference. This work paves the way for future research on efficient deepfake detection. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps keep online content trustworthy by stopping fake images from spreading quickly. It develops a new way to detect these fake images using special computer programs called Binary Neural Networks (BNNs). These BNNs are fast and accurate, making them better than previous methods that used big models. The team tested this approach on several datasets and found it worked well in most cases, with some extra benefits like being able to process information 20 times faster. |
Keywords
» Artificial intelligence » Inference
