book

Deepfake and Image Forgery Detection

Name: Deepfake and Image Forgery Detection
Author: Chad Johnson
ISBN: 9783111734330

by Chad Johnson

April 2026

Intermediate

463 pages

17h 9m

English

De Gruyter

Read now

Unlock full access

Frontmatter
Acknowledgments
Contents
1 Introduction to Deepfakes and Image Forgeries
1.1 Definition and Scope1.1.1 A Taxonomy of Images1.2 Historical Context and Evolution1.3 Impact on Society and Media1.3.1 Political Manipulation in Practice1.3.2 Celebrity and Public Figure Impersonation1.3.3 Financial Fraud and Security Breaches1.3.4 Erosion of Public Trust1.3.5 Regulatory and Policy Response1.3.6 Economic Costs and Emerging Markets1.3.7 Legal Evidence and Courtroom Admissibility1.3.8 Child Safety Implications1.3.9 Mental Health Impact on Victims1.3.10 Disinformation in Conflict Zones1.3.11 Journalistic Verification Workflows1.3.12 Educational Consequences1.3.13 Cultural Heritage and Art Market Fraud1.4 Foundational Signals: What They Are and Why They Matter1.5 Challenges and the Need for Detection1.6 How to Use This Book1.6.1 Why This Book Exists1.6.2 What This Book Covers—and What It Doesn’t1.6.3 Foundational Ideas You’ll Meet Immediately1.6.4 How to Use This Book1.6.5 Learning Outcomes1.6.6 The Map: From Physics to Testimony1.6.7 Standards and Admissibility1.6.8 Ethical Stance1.6.9 Prerequisites and Study Advice1.6.10 A Final Word on Mindset1.7 Summary of Terms
2 Principles of Multimedia Forensics
2.1 Foundational Concepts2.1.1 Definition and Key Objectives2.2 Aspects of an Image That Contains Forensic Artifacts2.2.1 Substantive Qualities2.2.2 Contextual Qualities2.3 PRNU (Photo-Response Nonuniformity): A Plain-Language Primer2.3.1 How Analysts Use PRNU2.3.2 What PRNU Is—and Isn’t2.3.3 When PRNU Gets Tricky2.3.4 Can PRNU Itself Be Forged or Removed?2.3.5 Where PRNU Comes from in the Imaging Chain2.3.6 How We Explain a PRNU Finding2.4 JPEG Foundations: DCT Blocks and Quantization2.5 Recompression and Double Compression2.6 The Discrete Cosine Transform (DCT)2.7 Quantization and Quantization Tables2.8 Chroma Subsampling (4:4:4, 4:2:2, 4:2:0)2.9 MCUs and the JPEG Grid2.10 Entropy Coding and Marker Fingerprints2.11 RGB↔YCbCr and Transfer Functions (Gamma)2.12 Resampling and Interpolation Kernels2.13 Demosaicing and CFA Patterns2.14 Rolling Shutter and Temporal Sampling2.15 Perceptual Hashing Versus Cryptographic Hashing2.16 “ELA” and Transform-Domain Thinking2.17 JPEG Versus HEIC/AVIF (Modern Codecs)2.18 Dynamic Range, Bit Depth, and Banding2.19 Conclusion2.20 Summary of Terms2.20.1 Small Clusters (Taxonomy/Axes)
3 Methods of Image Manipulation
3.1 Types of Digital Manipulation3.2 Pixel-Level Alterations3.2.1 Brightness, Contrast, and Color Correction3.2.2 Local Tone Mapping3.2.3 Gamma Adjustment3.2.4 Blurring, Sharpening, and Noise Insertion/Removal3.2.5 Stylization3.2.6 Local Region Enhancements (e.g., Teeth Whitening and Skin Smoothing)3.3 Geometric Transformations3.3.1 Rotation, Scaling, and Skewing3.3.2 Cropping and Reframing3.3.3 Flipping or Mirroring3.3.4 Lens Correction and Intentional Distortion3.3.5 Resampling Artifacts3.3.6 A Practical Examination Flow3.4 Copy-Move Forgery (Clone Stamping)3.4.1 Patch-Based Duplication3.4.2 Texture Cloning3.4.3 Region-Based Detection3.4.4 Keypoint-Based Block Matching3.4.5 Block-Based Self-Similarity Detection3.4.6 Dense Field Detection and Offset Analysis3.4.7 Region Growing and Transformation Clustering3.4.8 Clone Inconsistency in Texture, Noise, or Geometry3.5 Splicing and Compositing3.5.1 Hard- and Soft-Edge Blending3.5.2 Shadow and Lighting Inconsistencies3.5.3 Focus and Depth-of-Field Mismatch3.5.4 Noise and Compression Inconsistencies3.6 Traditional Editing and Retouching3.6.1 Healing Brush Artifacts3.6.2 Airbrushing and Smoothing Tools3.6.3 Edge Enhancement Exaggeration3.6.4 Color, White Balance, and Gamut3.6.5 Reflections, Occlusions, and Contact Geometry3.6.6 Resampling and Matting Signatures3.7 AI-Based and Synthesis Manipulations3.7.1 AI-Generated and GAN-Based Forgeries3.7.2 Deepfake Facial Reenactment3.7.3 Entire Image Generation (StyleGAN, DALL · E, and Midjourney)3.7.4 Face Morphing, Age Progression, Expression Transfer3.7.5 Detection by GAN Fingerprinting, Color Statistics, or Reverse Model Attribution3.8 Anti-forensic Techniques3.8.1 Anti-forensics3.8.2 Metadata Removal, Spoofing, or Replacement3.8.3 Histogram Equalization to Conceal Tampering3.8.4 Recompression to Erase Copy-Move Footprints3.8.5 Intentional Noise or Watermark Disruption3.9 Hybrid Manipulations3.9.1 Hybrid Techniques3.9.2 Multi-method Integration3.9.3 Intentional Overlap of Methods3.9.4 Layered Compositing, Retouching, and GAN Polishing3.10 Video-Specific Manipulations3.10.1 Temporal Tampering3.10.2 Inter-frame Forgery3.10.3 Intra-frame Manipulation3.10.4 Speed and Duration Manipulation3.10.5 Audio Manipulation and Mismatch3.10.6 Compression Artifacts as Indicators3.11 Summary of Terms
4 Best Practices and Forensic Procedure
4.1 A Standard-Driven Foundation4.1.1 Admissibility and Gatekeeping (Daubert, FRE 702, and Beyond)4.1.2 Validation Artifacts (Produce These on Every Case)4.1.3 Reporting Artifacts and Document Types4.2 Evidence Preservation and Chain of Custody in Digital Media4.2.1 Authenticity and Integrity4.2.2 Chain of Custody4.2.3 Validity Without Matching Hashes4.2.4 Secure Storage and Non-repudiation4.2.5 Closing Observations4.3 Scientific Methodologies4.3.1 Analytical Frameworks in Forensic Investigations4.3.2 Comparison Between Traditional and Digital Forensics4.4 Structured Examination4.5 Interpretation, Reporting, and Testimony4.6 Reporting Language and Legal Gatekeeping4.7 Archiving4.8 Bias, Standards, and Provenance in Practice4.9 Tools and Technologies4.9.1 Blind Proficiency as Routine, Not Event4.9.2 Display Fidelity and Color Management4.10 Summary of Terms
5 Methods of Image Authentication
5.1 Active Authentication: Preventing Ambiguity Before It Begins5.1.1 Substantive and Contextual Analysis5.1.2 Feature-Based and Statistical Analysis5.1.3 Compression and Encoding Forensics5.1.4 Sensor and Optical Signature Analysis5.1.5 Motion and Temporal Analysis5.2 Integrating the Methods5.3 Methods of Image Authentication5.4 Overview of Authentication Techniques5.4.1 Digital Signatures5.4.2 Watermarks5.4.3 Cryptographic Hash Functions5.4.4 Blockchain-Based Provenance Systems5.5 Comparison of Traditional and Modern Approaches5.6 Algorithmic Approaches5.6.1 Machine Learning, Statistical Models, and Analytical Detection5.6.2 Machine Learning Models5.6.3 Statistical Inference and Signal Analysis5.6.4 Error Level Analysis and Visual Consistency Testing5.7 Challenges and Future Prospects5.7.1 Limitations of Contemporary Techniques and the Road Ahead5.7.2 Vulnerability to Sophisticated Attacks5.7.3 False Positives and Negatives5.7.4 Scale, Speed, and Resource Constraints5.8 Directions for Enhancing Robustness5.8.1 Architectures, Adaptation, and the Next Generation of Tools5.8.2 Hybrid Systems5.8.3 Adaptive Algorithms and Continuous Learning5.8.4 Emerging Technologies and Infrastructure5.8.5 Standardization and Collaborative Development5.9 Conclusion5.10 Summary of Terms
6 Substantive and Contextual Analysis
6.1 Surface-Level Visual Anomalies: Perceptual Coherence6.1.1 Lighting and Shadow Inconsistencies6.1.2 Color Temperature and White Balance6.1.3 Depth of Field and Focus Mismatch6.1.4 Compression Artifacts and Resolution Mismatch6.1.5 Noise and Grain Inconsistencies6.2 Spatial Integration and Physical Plausibility6.2.1 Perspective and Spatial Relationships6.3 Rooting and Contact Physics6.3.1 Seaming and Webbing6.3.2 Environmental Inconsistencies6.3.3 Common Cues and Contradictions6.3.4 Analytical Techniques6.4 Biological and Anatomical Cues6.4.1 Anatomical and Biological Inconsistencies6.4.2 Typical Biological Errors6.4.3 Techniques and Tools6.4.4 Specular Highlights and Reflections6.4.5 Emotion and Expression Cues6.5 Symbolic and Cultural Plausibility6.5.1 Fashion, Technology, and Cultural Anachronisms6.5.2 Common Anachronisms and Cultural Incongruities6.5.3 Text and Symbol Errors6.5.4 Contextual Plausibility and Narrative Coherence6.6 Metadata, Provenance, and Contradictions6.6.1 Metadata–Contradictory Content6.7 Summary of Terms
7 Feature-Based Analysis: Extraction and Identification
7.1 Hashing and Robust Feature Matching7.1.1 Cryptographic Hashes (Fragile)7.1.2 Perceptual Hashing (pHash, aHash, and dHash)7.1.3 Robust Local Features (SIFT, SURF, and ORB)7.1.4 Global Descriptors and Learned Embeddings7.1.5 Applications in Forensics7.1.6 Limitations and Best Practices7.2 Frequency-Domain Analysis7.2.1 Frequency-Domain Analysis: Seeing Beyond the Pixels7.3 Audio Consistency Checks7.4 Texture Analysis7.5 Texture Analysis for Image Forgery Detection7.5.1 What Is Texture in a Forensic Context?7.6 Texture Coherence and Internal Consistency7.6.1 Common Texture Anomalies in Forged Media7.6.2 Tools and Techniques for Texture Analysis7.6.3 Integrating Texture Analysis with Other Forensic Methods7.7 Statistical Methods for Texture Characterization7.7.1 Tools7.8 Copy-Move Detection and Self-Similarity in Texture7.9 Keypoint Descriptor Comparison in Image Forgery Detection7.9.1 What Are Keypoints and Descriptors?7.9.2 SIFT (Scale-Invariant Feature Transform)7.9.3 SURF (Speeded-Up Robust Features)7.9.4 ORB (Oriented FAST and Rotated BRIEF)7.9.5 Descriptor Matching Techniques7.9.6 Geometric Consistency Verification7.9.7 Visualizing and Validating Matches7.10 Edge and Contour Consistency in Forgery Detection7.10.1 Understanding Edge Semantics7.10.2 Forensic Cues Derived from Edge Analysis7.10.3 Algorithmic Edge Analysis7.10.4 Application in Copy-Move and Splicing Detection7.10.5 Visual Acuity and Training7.11 Deep Learning and Learned Texture Models7.11.1 Texture Representation in Deep Neural Networks7.11.2 Detecting GAN-Generated Texture Anomalies7.11.3 GAN Fingerprinting with Texture Signatures7.11.4 Applications and Integration7.12 Texture-Based Segmentation and Localization7.12.1 Understanding Texture-Based Segmentation7.12.2 Techniques for Localization7.12.3 Visualization and Human Interpretation7.12.4 Integration with Other Modalities7.13 Texture Transfer and Stylization Artifacts7.14 Manual Texture Inspection Techniques7.15 Summary of Terms

8 Compression and Encoding Forensics
8.1 DCT Coefficient Mapping and Quantization Signatures8.2 JPEG Artifacts and Noise Analysis8.3 Understanding JPEG Compression8.3.1 Double JPEG Compression and Grid Misalignment8.4 Noise Profiles and Sensor Fingerprints8.5 Differentiating Real from GAN-Based Images8.6 Practical Considerations and Caveats8.7 Digital Noise: Structure, Sources, and Forensic Value8.7.1 Types of Noise and Their Origins8.7.2 Camera-Specific Noise Patterns8.7.3 Photographic Practices That Modify Noise8.7.4 Noise Discontinuities and Tampering Detection8.7.5 Limitations and Considerations8.8 Error Level Analysis (ELA)8.9 Double-Compression Detection8.10 Double JPEG Compression Artifacts8.10.1 How JPEG Compression Works8.10.2 What Happens in Double Compression8.10.3 Forensic Detection Techniques8.10.4 Practical Considerations and Limitations8.10.5 Applications8.11 Camera Signature-Based Detection8.11.1 Physical Signatures: Sensor Pattern Noise (SPN) and PRNU8.11.2 Processing Signatures: CFA, Demosaicing, and Compression Artifacts8.11.3 Source Camera Attribution and Anti-forensics8.12 CFA Pattern Analysis and Demosaicing Consistency8.12.1 The Bayer Pattern and Color Subsampling8.12.2 Detection Techniques and Analytical Tools8.12.3 Common Forensic Scenarios8.12.4 Practical Considerations and Limitations8.13 Chromatic Aberration Analysis8.13.1 Forensic Use of Chromatic Aberration8.14 Lens Distortion Patterns8.14.1 Forensic Application8.14.2 Integration with Other Analyses8.15 JPEG Ghost Detection8.15.1 Principle of JPEG Ghost Detection8.15.2 Operational Methodology8.15.3 Analytical Interpretation8.15.4 Technical Considerations8.15.5 Applications8.15.6 Tool Support8.16 Quantization Table Analysis8.16.1 Understanding Quantization Tables in JPEG Compression8.16.2 Quantization Table as Forensic Evidence8.16.3 Applications in Forgery Detection8.17 Quantization Table Fingerprinting8.17.1 Quantization Table Tampering and Anti-forensics8.17.2 Analytical Techniques8.17.3 Limitations and Considerations8.18 Conclusion8.19 Summary of Terms
9 Sensor and Optical Signature Analysis
9.1 Sensor Fingerprinting: PRNU Without the Mystery9.1.1 Forensic Value9.1.2 Getting the Fingerprint (Step by Step)9.1.3 What PRNU Is—and Where It Comes from9.1.4 Matching a Questioned Image9.1.5 Localizing Tampering9.1.6 Failure Modes, Confounders, and Modern Pipelines9.1.7 Security Considerations: Copy Attacks and Defenses9.1.8 Reporting and Courtroom Framing9.2 CFA and Demosaicing Signatures9.2.1 Why CFA Traces Exist (and Why They Survive)9.2.2 What to Look for (Observable Signatures)9.2.3 Core Analytics (How to Quantify)9.2.4 A Lab-Ready Workflow (with Thresholds and Exhibits)9.2.5 Handling Real-World Complications9.2.6 Integrating with Resampling and Copy-Move Detectors9.2.7 Validation, Thresholds, and Reporting9.2.8 Quick-Reference Checklist (CFA/Demosaicing)9.3 Optical System Signatures: Vignetting, Distortion, Chromatic Aberration, PSF/Bokeh9.3.1 Why Optical Traces Matter9.3.2 Vignetting (Radial Falloff)9.3.3 Geometric Distortion (Barrel, Pincushion, and Moustache)9.3.4 Chromatic Aberration (CA) Maps9.3.5 Spatially Varying PSF and Bokeh9.3.6 A Unified Optical-Forensic Workflow9.3.7 Real-World Complications (and How to Handle Them)9.3.8 Integration with PRNU and CFA9.3.9 Validation and Thresholds9.3.10 Quick Checklist (Optics)9.4 Readout and Shutter Artifacts: Rolling Shutter, Row/Column Noise, Mains Flicker9.4.1 Why Readout Artifacts Matter9.4.2 Rolling Shutter: Geometry and Observables9.4.3 Estimating Rolling-Shutter Parameters in Practice9.4.4 Using Rolling-Shutter Coherence to Find Splices9.4.5 Mains-Rate Flicker: Phase and Stripe Analysis9.4.6 Row/Column Fixed-Pattern Noise (FPN) and Banding9.4.7 Practical Workflow (Video and High-Frame-Rate Stills)9.4.8 What Adversaries Do—and Counters9.4.9 Validation and Reporting9.5 Computational Photography: When Phones Rewrite the Signal9.5.1 Why This Matters for Forensics9.5.2 What Contemporary Pipelines Actually Do to the Signal9.5.3 Recognizing Computational Processing in Practice9.5.4 Adapting Attribution and Tamper Localization9.5.5 Distinguishing ISP Corrections from Human Edits9.5.6 Super-Resolution and Learned Hallucination as a Confound and a Clue9.5.7 Night-Mode Specifics9.5.8 Anti-forensics That Exploit These Pipelines—and Counters9.5.9 Validation, Thresholds, and Report Language9.6 PRNU for Video9.7 Putting It Together: A Practical Recipe9.8 How to Talk About This Under Daubert/FRE 7029.9 Color Filter Array (CFA) and Demosaicing Signatures9.9.1 What Exactly Survives the Pipeline9.9.2 How to Measure It Defensibly9.9.3 What Splices and Resampling Do to These Cues9.9.4 Non-Bayer Layouts and Modern Phone Pipelines9.9.5 Failure Modes and How to Avoid Them9.9.6 A Compact Workflow You Can Defend9.10 Optical System Signatures: Vignetting, Distortion, CA, PSF9.11 Readout and Shutter Artifacts: Rolling Shutter, Row/Column Banding, Flicker, Defect Constellations9.11.1 Rolling-Shutter Geometry You Can Measure9.11.2 Row/Column Pattern Noise and Fixed-Frequency Banding9.11.3 Illumination Flicker and Row-Time Aliasing (Video)9.11.4 A Defensible Workflow9.11.5 Sensor Defect Constellations9.11.6 Limits and Integration9.12 Illumination and Flash Signatures Tied to the Optics and the Sensor9.12.1 Radial Falloff Under On-Camera Lighting9.12.2 Catchlights and Specular Geometry in Eyes and Mirrors9.12.3 Spectral Fingerprints: IR-Cut, Color Pipeline, and White Balance9.12.4 Bringing the Cues Together in Analysis9.12.5 Confounders and Limits9.13 Computational Photography Effects (and How to Handle Them)9.13.1 What These Pipelines Do to Sensor and Optical Traces9.13.2 Recognizing Computational Footprints in Practice9.13.3 Handling Strategies: Adapting Tests and Reporting Limits9.13.4 A Short Decision Pattern for Examiners9.14 Video PRNU and Cross-Modal Linkage9.15 Anti-forensics, Validation, and Calibrated Conclusions9.16 Standard Operating Procedure: Building and Using a PRNU System9.16.1 Scope and Preconditions9.16.2 Reference Construction (Device Fingerprint)9.16.3 Operating Point and Sanity Panel9.16.4 Case Ingestion and Global Attribution9.16.5 Localization (Tamper Mapping)9.16.6 Anti-forensic Screening9.16.7 Corroboration Outside the Sensor Domain9.16.8 Quality Assurance, Reproducibility, and Change Control9.16.9 Reporting and One-Page Validation Sheet9.17 Summary of Terms
10 Motion and Temporal Analysis
10.1 Optical Flow Analysis10.1.1 Theoretical Foundations10.1.2 Core Algorithms10.2 Forensic Applications of Optical Flow10.2.1 Visualizing and Interpreting Flow10.2.2 Method Limits and the Error Budget10.2.3 Tools That Hold Up in Practice10.3 Temporal Inconsistencies10.3.1 Replay Attacks (Screen-on-Camera)10.3.2 Two- Versus Three-Dimensional Face Cues10.3.3 Physiological Signals (rPPG)10.3.4 Challenge-Response (When You Control Capture)10.3.5 Deepfake Video Liveness10.4 Liveness Detection10.4.1 Container and Bitstream Timeline10.4.2 Frame-Level Continuity10.4.3 Scene-Level Time Cues10.4.4 Cross-Modal Timing (A/V Sync)10.5 Court-Ready Workflow (Motion and Temporal)10.5.1 Motion Inconsistency Detection10.5.2 Scene-Flow Discontinuities (Global vs. Local Motion)10.5.3 Temporal Tampering (Frame Insertion/Deletion/Retiming)10.5.4 Deepfake and Facial Reenactment Cues10.5.5 Stabilization Artifacts and Recompression Fingerprints10.6 Summary of Terms
11 Learned Detection and Hybrid Systems
11.1 Deepfake Detection Models11.2 Robust Features11.3 Sparse Representation11.4 Dimensionality Reduction11.5 Dynamic Thresholding11.6 Putting It Together: Hybrid Systems You Can Defend11.7 Adversarial and Anti-forensic Pressure (and Defenses)11.8 Provenance and Watermark Signals (Complements, Not Substitutes)11.9 Benchmarking, Generalization, and Reporting for Learned Systems11.10 Summary of Terms
12 Legal and Ethical Considerations
12.1 Expression, Privacy, and the Democratic Commons12.1.1 Access, ToS, and Anti-circumvention12.2 Torts, Remedies, and Civil Exposure12.2.1 Spoliation and Legal Holds12.3 Evidence Law and Admissibility12.3.1 Rule 702 (2023 Amendments), Frye, and Kumho12.3.2 Self-Authenticating Electronic Records and Summaries12.4 Provenance Tech, Platforms, and the “Liar’s Dividend”12.5 Children, CSAI, and Mandatory Reporting12.6 Professional Duty: Rigor with Restraint12.6.1 Testifying Versus Consulting Expert12.6.2 Language Discipline12.7 Data Protection, Biometrics, and Consent12.7.1 Cross-Border Transfers12.8 Research Ethics, Datasets, and Dual Use12.9 Work with Journalists and Civil Society12.10 Operational Security and Practitioner Safety12.11 Putting It Together12.12 Summary of Terms
13 Future Directions in Image Forgery Detection
13.1 Diffusion-Native Detection and Open-World Robustness13.2 Cryptographically Anchored Provenance13.3 Watermarking That Survives the Wild13.4 Neuromorphic and Secure-Capture Hardware13.5 Federated and Privacy-Preserving Learning13.6 Real-Time, Edge-Native Verification13.7 Holistic, Multimodal Forensics13.8 Explainability, Calibration, and Legal Readiness13.9 A Convergent Outlook13.10 Summary of Terms
14 Glossary
SourcesAppendix A: Case Report Template
Index
Index

Content preview from Deepfake and Image Forgery Detection

11 Learned Detection and Hybrid Systems

Detecting synthetic or manipulated media with machine learning is not a single trick so much as a layered practice. You are balancing two pressures at once: on one side, ever-improving generators that erase the visual tells you once relied on; on the other, courts and review boards that demand methods you can explain, validate, and reproduce. This chapter shows how to weld modern “learned” models to classical, physics-guided forensics so you get the best of both: sensitivity in the wild and conclusions you can defend (1; 2). Throughout, detailed sensor signatures, codec and quantization fingerprints, and temporal mechanics are treated in the chapters dedicated to those topics; here, they are inputs ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

The Hidden Cost of AI-Assisted Creativity

Publisher Resources

ISBN: 9783111734330

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Deepfake and Image Forgery Detection

by Chad Johnson

11 Learned Detection and Hybrid Systems

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.