All classes Occlusion Robust Face Recognition Syllabus · Episode 1 of 6

▶ Video Lesson

Watch every step above performed live, with English narration.

0 plays

Class book

Page 1 / 7

Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand): Cover: Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Glasses, Hand)

Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand): Before you start, steps list, and Step 1

Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand): Step 2: Loading LFW and Data Split

Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand): Step 3: Exploring Number of Photos per Person

Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand): Step 4: Landmark-based Synthetic Occlusion

Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand): Step 5: Original face vs occluded face

Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand): Done: Final Result and Further Practice

The end

Cover: Face Recognition with Occlusion Tolerance #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Glasses, Hand)

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 #1: LFW Face Dataset & Creating Synthetic Occlusion (Mask, Sunglasses, Hand)”.

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

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 9 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 66 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 62 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

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

NSGA-II MOPSO Hybrid Multi-Objective Optimization

6 episodesFree preview

Python · Intermediate–Advanced

NSGA-II MOPSO Hybrid Multi-Objective Optimization

6 video lessons with PDF guides, 59 min in total. Starts with: Multi-Objective Optimization: Pareto Front, Dominance, and Test Functions ZDT/DTLZ.

59 min 9 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

Here is the final result, faces from the LFW dataset automatically covered with masks, sunglasses, and hands. Occlusion like this is the main enemy of face recognition systems, and in this series, we build a model that still recognizes faces even when they are covered. Let's get started. Hello, welcome to the AI research tutorial series, face recognition with occlusion resistance using multi-scale feature fusion. Face-based attendance and security systems often fail when faces are covered by masks, sunglasses, or hands. This condition is called occlusion. In this 6-episode series, we build a complete system, dataset and synthetic occlusion, face detection and alignment, feature pyramid, multi-scale fusion with arc face loss, verification with ROC curve, up to occlusion resistance testing and face attendance application. In this first episode, we prepare the data.

Step 1: Installing the library & LFW dataset

We open the terminal through the terminal menu, new terminal, then install the libraries for this series, Torque, Torchvision, OpenPithon, Scikit-learn, Medplotlib, and Streamlit. The dataset is label Faces in the Wild or LFW, a collection of photos of public figures from news. Scikit-learn can download it automatically, and the version for the customer is available in the special project file for premium customers. Create a file named data_underscore_lfw.py

Step 2: Loading LFW and splitting the data

The load function takes people who have at least 20 photos, in color and original resolution, then saves them as a NPZ file so that the next loading is instant. The split data function divides the data into 70, 15, 15% stratified per person, so that each person appears in the training, validation, and test data.

Step 3: Exploring the number of photos per person

Now create the file exploration_underscore_lfw.py and run it. There are 3,023 photos of 62 people, each 125x94 pixels. The number of photos is highly unbalanced, with the minimum being 20, the median 31, and George Wibu alone has 530 photos. This imbalance must be remembered when evaluating accuracy. The split is 2,116 training, 453 validation, and 454 test. All of them contain 62 people.

Step 4: Landmark-based synthetic occlusion

Create a file named occlusion.py Synthetic occlusion is placed based on 5 facial landmarks, 2 eyes, nose, and 2 corners of the mouth, so that the position adjusts according to the face size. The mask is a polygon from below the nose to the chin with 3 color options, sunglasses are two ellipses above the eyes along with the bridge, and the hand is an ellipse in skin color taken from the median color of the face itself, with random position and angle. Return to the exploration file.

Step 5: Original face vs. occluded face

We import the add occlusion function. Then, for the raw LFW photos, we use the average landmark position estimate, in the next episode, landmarks are detected for each face. Run. The first graph shows the imbalance in the number of photos per person. The second graph shows 4 faces in 4 versions. Original, mask, sunglasses, and hand. In this way, we can create as many occluded faces as needed without retaking photos, for augmentation during training as well as for testing.

Done!

Done. We have now learned about occlusion problems in face recognition, created the LFW dataset with 62 people, split the data in a stratified manner, and created synthetic occlusions for masks, sunglasses, and hands. In the next episode, we will detect facial landmarks with units, and align each face to the ArcV 112 pixel template. Currently working on a thesis, dissertation, or AI research in the field of face recognition and biometrics? Cerdas Inovasi Digital helps with deep learning implementation, from synthetic occlusion, face alignment, pyramid features, art face, verification, and ROC, all the way to face attendance applications, as well as opening deep learning courses with PyTor. Want to master the entire series? From face datasets to attendance applications? All 6 episodes Complete with a step-by-step PDF guide and source code, 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 single classes with lifetime access. Subscribe now on the premium page cerdasinovasidigital.com Contact our WhatsApp, Or visit cerdasinovasidigital.com Thank you for watching, See you in the next episode. Thank you for watching,

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

1:02Step 1: Installing the library & LFW dataset

1:34Step 2: Loading LFW and splitting the data

3:43Step 3: Exploring the number of photos per person

6:09Step 4: Landmark-based synthetic occlusion

9:50Step 5: Original face vs. occluded face

12:12Done!

Ask the class material

My notes

Questions & answers

Your instructor

▶ Video Lesson

Recommended classes for you

Need more than a video?

Transcript

0:03Introduction

1:02Step 1: Installing the library & LFW dataset

1:34Step 2: Loading LFW and splitting the data

3:43Step 3: Exploring the number of photos per person

6:09Step 4: Landmark-based synthetic occlusion

9:50Step 5: Original face vs. occluded face

12:12Done!

Ask the class material

My notes

Questions & answers

Your instructor

Introduction

Step 1: Installing the library & LFW dataset

Step 2: Loading LFW and splitting the data

Step 3: Exploring the number of photos per person

Step 4: Landmark-based synthetic occlusion

Step 5: Original face vs. occluded face

Done!

Introduction

Step 1: Installing the library & LFW dataset

Step 2: Loading LFW and splitting the data

Step 3: Exploring the number of photos per person

Step 4: Landmark-based synthetic occlusion

Step 5: Original face vs. occluded face

Done!