Build a Fast Deepfake Detector in an Afternoon (2026)

A developer published a step-by-step guide for building a production-ready deepfake detector in an afternoon, combining a frozen MobileNet-V3 Small backbone with a temporal attention head and packaging the model in a Dockerized Flask API. The writeup covers dataset acquisition from DFDC-2023, FaceForensics++ and Kaggle, a preprocessing script that extracts 2-second clips at 15 fps, and a CPU fallback via ONNX runtime, with reported throughput of 12 fps on an RTX 3060 and 1 fps on an i7-12700 CPU.

Every day you hear about a “viral” video that looks real but turns out to be AI‑fabricated. The truth? You can spot most of them with a tool you build in an afternoon. This guide walks you through a complete, production‑ready deepfake detector—from data collection to a Docker‑ized service you can run on a laptop or a cloud GPU. By the end you’ll have: No PhD required—just a terminal, a modest GPU, and a curiosity to stop synthetic media from hijacking elections, scams, and misinformation campaigns. | Step | Command / Code | What it does | |---|---|---| | 1️⃣ Clone the repo | git clone https://github.com/yourname/deepfake-detector.git && cd deepfake-detector | Pulls the starter code and Dockerfile. | | 2️⃣ Build the Docker image | docker build -t deepfake-detector:latest . | Packages the model, dependencies, and a tiny Flask API. | | 3️⃣ Run the container | docker run -p 8080:8080 deepfake-detector:latest | Exposes a local endpoint at http://localhost:8080/predict . | | 4️⃣ Test with a video | curl -X POST -F "file=@sample.mp4" http://localhost:8080/predict | Returns JSON with score , heatmap url , and model version . | | 5️⃣ Deploy to the cloud optional | docker tag deepfake-detector:latest yourregistry/deepfake-detector:1.0 && docker push yourregistry/deepfake-detector:1.0 | Pushes the image to a container registry for AWS ECS, GKE, etc. | | Device | VRAM | Real‑time FPS 30 s clip | Notes | |---|---|---|---| | RTX 3060 8 GB | 8 GB | 12 fps MobileNet‑V3 | Good balance of cost and speed. | | RTX 2070 8 GB | 8 GB | 9 fps EfficientNet‑B0 | Slightly slower, still usable. | | CPU‑only i7‑12700 | – | 1 fps MobileNet‑V3 | OK for batch jobs, not live streaming. | If you only have a CPU, set the environment variable USE CPU=1 before launching the container; the code will automatically fall back to the ONNX runtime with CPU execution providers. Create a clean Python env python3 -m venv venv && source venv/bin/activate Install core libraries pip install -U pip setuptools wheel pip install torch==2.2.0 torchvision==0.17.0 onnxruntime==1.16.0 \ opencv-python==4.9.0 ffmpeg-python==0.2.0 \ flask==3.0.0 tqdm==4.66.1 pandas==2.2.1 Tip: Use torch.cuda.is available inside a Python REPL to verify GPU access before proceeding. | Source | Type | How to download one‑liner | |---|---|---| | DFDC‑2023 | 1 M videos real + fake | wget -O dfdc2023.zip https://research.google.com/dfdc/dfdc2023.zip && unzip dfdc2023.zip | | FaceForensics++ | High‑quality face swaps | git clone https://github.com/ondyari/FaceForensics.git && cd FaceForensics && ./download.sh | | DeepFakeDetectionChallenge Kaggle | 50 k labeled clips | kaggle competitions download -c deepfake-detection-challenge | After downloading, run the preprocessing script to extract 2‑second clips and generate frame‑level labels: python scripts/preprocess.py \ --input-dir /data/dfdc2023 \ --output-dir /data/processed \ --clip-length 2 \ --frame-rate 15 The script also creates a CSV manifest manifest.csv that the training pipeline expects. We selected MobileNet‑V3 Small as the backbone because: The detection head is a temporal attention module that aggregates per‑frame embeddings into a single video‑level score. python import torch, torch.nn as nn, torch.optim as optim from torchvision import models, transforms from dataset import DeepFakeDataset 1️⃣ Load backbone backbone = models.mobilenet v3 small pretrained=True .features backbone.eval freeze ImageNet weights 2️⃣ Temporal attention head class AttnHead nn.Module : def init self, dim=576 : super . init self.attn = nn.MultiheadAttention dim, num heads=4 self.fc = nn.Linear dim, 1 def forward self, x : x: T, B, C attn out, = self.attn x, x, x pooled = attn out.mean dim=0 B, C return torch.sigmoid self.fc pooled model = nn.Sequential backbone, nn.AdaptiveAvgPool2d 1 , nn.Flatten , AttnHead model = model.cuda 3️⃣ Optimizer & loss criterion = nn.BCELoss optimizer = optim.AdamW model.parameters , lr=1e-4, weight decay=1e-5 4️⃣ Training loop for epoch in range 5 : for clips, labels in DataLoader DeepFakeDataset ... , batch size=8, shuffle=True : clips = clips.cuda ; labels = labels.cuda .float preds = model clips .squeeze loss = criterion preds, labels optimizer.zero grad ; loss.backward ; optimizer.step print f"Epoch {epoch} – loss {loss.item :.4f}" Result: After 5 epochs on a single RTX 3060 we reached 84 % AUC on the held‑out DFDC‑2023 test set. python scripts/export onnx.py \ --checkpoint checkpoints/best.pt \ --output model.onnx \ --opset 17 The exported model runs at ~25 fps on an RTX 3060 using onnxruntime-gpu . python python app.py import io, json, torch, onnxruntime as ort, cv2, numpy as np from flask import Flask, request, jsonify app = Flask name session = ort.InferenceSession "/app/model.onnx", providers= "CUDAExecutionProvider" def video to tensor path : cap = cv2.VideoCapture path frames = while len frames < 30: 2‑second clip @ 15 fps ret, frame = cap.read if not ret: break frame = cv2.resize frame, 224, 224 frames.append frame.transpose 2,0,1 cap.release return np.stack frames .astype np.float32 / 255.0 @app.route "/predict", methods= "POST" def predict : file = request.files "file" tmp = "/tmp/input.mp4" file.save tmp tensor = video to tensor tmp None, ... 1, T, C, H, W tensor = tensor.transpose 0,2,1,3,4 1, C, T, H, W for ONNX outs = session.run None, {"input": tensor} 0 score = float outs.squeeze return jsonify {"deepfake score": score, "model version": "mobilev3‑attn‑v1"} if name == " main ": app.run host="0.0.0.0", port= --- Herramienta mencionada: GitHub Copilot https://github.com/features/copilot