All classes Occlusion Robust Face Recognition Syllabus · Episode 5 of 6

Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER

0 views

Face verification tutorial: cosine similarity, threshold, ROC curve, AUC, and Equal Error Rate (EER) using PyTorch and scikit-learn. Episode 5 of the face recognition with occlusion resistance series.

Software: Deep Learning (PyTorch ArcFace)
Level: Advanced
Duration: ± 9 min
Series total: 1 h 26 min · 6 videos
Published: October 24, 2026

Episode 5 of 6 · view syllabus

Download PDF (5 PDF pages)

0 downloads

Download Project File· 153 MB Unlock All Premium Classes ▶ Watch the video lesson0 plays

▶ Video Lesson

Watch every step above performed live, with English narration.

0 plays

Class book

Page 1 / 5

Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER: Cover: Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER

Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER: Before you start, steps list, and Step 1

Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER: Step 2: ROC Curve, AUC & EER

Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER: Step 3: Score Distribution & ROC Curve

Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER: Done: Final Result and Further Practice

The end

Cover: Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER

Click a page or use ◀ ▶ to turn · swipe on mobile · to enlarge

Want to go deeper?

Join one of our courses or ask us directly on WhatsApp.

Download PDF· 0 downloads View the full 6-episode syllabus View Face Recognition Courses Ask on WhatsApp

Recommended classes for you

Picked from the software, topic, and level of “Face Recognition with Occlusion Tolerance #5: Face Verification: Cosine Similarity, Threshold, ROC Curve & EER”.

6 episodesFree preview

Python · Intermediate–Advanced

Deepfake Detection Vision Transformer

6 video lessons with PDF guides, 1 h 11 min in total. Starts with: Deepfake Concepts, Manipulation Types & Real vs Fake Face Dataset.

1 h 11 min 0 0

View class

18 episodesFree preview

Python · Beginner–Advanced

Machine Learning Python

18 video lessons with PDF guides, 3 h 41 min in total. Starts with: Introduction to Machine Learning & First Model.

3 h 41 min 61 0

View class

6 episodesFree preview

Python · Intermediate–Advanced

Brain Tumor Segmentation with U-Net

6 video lessons with PDF guides, 1 h 10 min in total. Starts with: Brain MRI Dataset & Tumor Mask: Loading, Visualization & Patient-wise Splitting.

1 h 10 min 28 0

View class

50 episodesFree preview

Agentic AI · Beginner–Intermediate

Agentic AI Without Coding

50 video lessons with PDF guides, 14 h 29 min in total. Starts with: Chatbots, Automation, and AI Agents: What's the Difference?.

14 h 29 min 74 0

View class

100 episodesFree preview

Claude · Beginner–Advanced

Claude from Zero to Expert

100 video lessons with PDF guides, 15 h 1 min in total. Starts with: What Claude Is and How It Differs from ChatGPT and Gemini.

15 h 1 min 69 0

View class

Deep Learning CNN-LSTM Pneumonia Detection

6 episodesFree preview

Python · Intermediate–Advanced

Deep Learning CNN-LSTM Pneumonia Detection

6 video lessons with PDF guides, 57 min in total. Starts with: Chest X-ray Dataset: Loading, Exploration & Data Splitting.

57 min 22 0

View class

6 episodesFree preview

Python · Intermediate–Advanced

Transfer Learning for Leaf Disease

6 video lessons with PDF guides, 1 h 5 min in total. Starts with: PlantVillage Dataset: Folder Structure, Class Distribution & Data Split.

1 h 5 min 10 0

View class

ANFIS Metaheuristics Time Series Forecasting

6 episodesFree preview

Python · Intermediate–Advanced

ANFIS Metaheuristics Time Series Forecasting

6 video lessons with PDF guides, 59 min in total. Starts with: Non-Linear Time Series: Dataset, Stationarity, and Patterns.

59 min 13 0

View class

Need more than a video?

Other ways we can help with Python.

Learn with a mentor

Scheduled AI & Machine Learning courses with live guidance and a certificate.

Course

Multi-Objective Optimization Course: NSGA-II, MOPSO & Hybrid Algorithms (Python)

Course

ECG Arrhythmia Classification Course with Wavelet & Deep Learning (PyTorch)

Course

ANFIS + PSO/GA Course for Non-Linear Time Series Forecasting (Python)

See all in AI & Machine Learning →

Have it built for you

No time to learn Python? Our team can do the project for you.

Service

Python Machine Learning — Prediction Models, Customer Segmentation, Fraud Detection & AI Apps

Service

AI Image Creation with Stable Diffusion — Unlimited Custom Visual Production

★ 4.9

Service

AI Image & Video Creation with Leonardo AI — All-in-One Visuals for Business

★ 4.9

See all in AI & Machine Learning →

Get notified about new Python classes

At most one email a week. Unfollow anytime with one click.

Transcript

The full narration of this video. The current sentence is highlighted while the video plays; click any sentence to jump there.

Show transcript

Introduction

This is the final result, more than 100,000 face pairs compared, and the score of the same person is clearly separated from the score of different people. From the ROC curve, we get an equal error rate around 1% and this helps make a decision for the verification system. Let's start. Hello, welcome to episode five of the facial recognition series with occlusion resistance. There are two main tasks in facial recognition. Identification answers who this person is, and verification answers whether these two photos are of the same person, like unlocking a phone with facial recognition. Verification requires a threshold, above that value is accepted, below it is rejected. Today we measure it using the standard biometric research method, the ROC curve and equal error rate.

Step 1: Cosine similarity for all pairs

Create verification.py file We load the best multi-scale fusion model from episode 4 and calculate the embedding for all test face data. Since the embedding is already normalized, matrix multiplication E times E transpose directly produces cosine similarity for every pair. We take the upper triangle of the matrix so that each pair is calculated once, then we mark pairs of the same person and different people. Run There are more than 102,000 pairs, around 5,000 same people and almost 98,000 different people. The average cosine of the same person is almost 0.9, while different people is around 0, meaning embeddings of different people are almost perpendicular.

Step 2: ROC Curve, AUC & EER

Now the ROC curve. For each threshold, the false accept rate is the percentage of different person pairs that are incorrectly accepted, and the true accept rate is the percentage of same person pairs that are correctly accepted. Equal error rate is the point when false acceptance equals false rejection, one concise number often reported in papers. Run AUC is around 0.999 and IIR is around 1%. Security systems usually use a stricter FAR, for example 0.1%, with a higher threshold. Threshold.IIR we save to threshold.json. Let's draw it.

Step 3: Score distribution & ROC curve

The left graph is the score distribution, blue for different people around 0, orange for same people clustered near 1, and the dashed line is the threshold. Very little overlap. The right ROC curve graph has a logarithmic FAR axis so that the small FAR region is clearly visible, and the red dot is IIR. Remember, the people in this test data are also in the training data, only their photos are different. For completely new people, the performance will be lower, and this will be addressed in the next episode.

Done!

Done We have distinguished between identification and verification, calculated cosine similarity for 100,000 pairs, created a ROC curve, calculated AUC and equal error rate, and selected the decision threshold. In the final episode, we tested the model's resistance to occlusion and built a facial attendance application. Currently working on a thesis, dissertation, or AI research in the field of facial recognition and biometrics? Cerdas Inovasi Digital helps implement deep learning, from synthetic occlusion, face alignment, feature pyramid, art face, verification and ROC, up to facial attendance applications, and also opens deep learning courses with PyTorch. Want to master the entire series, from face datasets to attendance applications? All 6 episodes, complete with PDF step-by-step guides and source code, are available for premium tutorial subscribers. Subscribe starting at Rp69.000 per month, save 28% with the 12-month package, or pay once Rp999.000 for lifetime access. Want just one class? Buy a single class with lifetime access. Subscribe now at the premium page cerdasinovasidigital.com Contact us via WhatsApp, or visit cerdasinovasidigital.com Thank you for watching, see you in the next episode.

Ask the class material

Type a question or keyword; we find the part of this episode that explains it and jump the video there.

My notes

Notes are saved with the current video time and are only visible to you.

Questions & answers

Stuck on a step? Ask here; the instructor and other learners can answer.

No questions yet.

Your instructor

Cerdas Inovasi Digital Team

Software developers, BIM & 3D engineers, and AI practitioners

53 video classes1,296 episodes

We build software, BIM models, games and AI automations for clients every day, and turn that hands-on work into step-by-step classes: every step is shown on screen and comes with a PDF guide you can keep.

See our clients →

Memuat

▶ Video Lesson

Recommended classes for you

Need more than a video?

Transcript

0:03Introduction

0:53Step 1: Cosine similarity for all pairs

3:28Step 2: ROC Curve, AUC & EER

4:55Step 3: Score distribution & ROC curve

7:21Done!

Ask the class material

My notes

Questions & answers

Your instructor

▶ Video Lesson

Recommended classes for you

Need more than a video?

Transcript

0:03Introduction

0:53Step 1: Cosine similarity for all pairs

3:28Step 2: ROC Curve, AUC & EER

4:55Step 3: Score distribution & ROC curve

7:21Done!

Ask the class material

My notes

Questions & answers

Your instructor

Introduction

Step 1: Cosine similarity for all pairs

Step 2: ROC Curve, AUC & EER

Step 3: Score distribution & ROC curve

Done!

Introduction

Step 1: Cosine similarity for all pairs

Step 2: ROC Curve, AUC & EER

Step 3: Score distribution & ROC curve

Done!