1. Which technique is commonly used for detecting human faces in images?
(A) Histogram Equalization
(B) Viola-Jones Algorithm
(C) Canny Edge Detection
(D) Fourier Transform
2. What is the primary role of Haar-like features in face detection?
(A) Compressing face data
(B) Representing face regions using simple patterns
(C) Enhancing face texture
(D) Reducing image noise
3. Which classifier is used in the Viola-Jones face detection framework?
(A) Naive Bayes
(B) AdaBoost
(C) SVM
(D) K-Means
4. Which of the following methods is based on projecting facial images onto a lower-dimensional space?
(A) DeepFace
(B) Eigenfaces
(C) LBP
(D) Gabor Filters
5. In PCA-based face recognition, what does each eigenface represent?
(A) A unique person
(B) An average face
(C) A feature vector
(D) A principal component of facial variation
6. What does LBP stand for in face recognition?
(A) Local Binary Pattern
(B) Linear Binary Projection
(C) Low Bandwidth Protocol
(D) Light Based Pattern
7. Which deep learning model was developed by Facebook for face recognition?
(A) FaceNet
(B) DeepID
(C) DeepFace
(D) OpenFace
8. Which distance metric is commonly used in face recognition to compare feature vectors?
(A) Euclidean Distance
(B) Manhattan Distance
(C) Mahalanobis Distance
(D) Hamming Distance
9. Which of the following is not a face recognition technique?
(A) LBP
(B) HOG
(C) K-means
(D) Eigenfaces
10. Which of these best describes the term “face embedding”?
(A) A filter used in preprocessing
(B) A set of pixel intensities
(C) A numerical vector representing a face
(D) A data compression technique
11. Which algorithm uses cascades of classifiers for face detection?
(A) CNN
(B) Viola-Jones
(C) PCA
(D) LDA
12. What is the key advantage of using CNNs in face recognition tasks?
(A) Faster computation
(B) Feature extraction from raw images
(C) Less training data required
(D) No need for labeled data
13. Which of the following techniques is most robust against variations in lighting for face recognition?
(A) Histogram Equalization
(B) PCA
(C) LBP
(D) Fourier Transform
14. In Deep Learning, which architecture is commonly used for facial recognition tasks?
(A) RNN
(B) LSTM
(C) CNN
(D) GAN
15. What is the function of a feature extractor in a face recognition system?
(A) Detect motion
(B) Compress image
(C) Identify facial landmarks
(D) Generate face descriptors
16. Which of the following is a standard benchmark dataset for evaluating face recognition algorithms?
(A) MNIST
(B) COCO
(C) LFW
(D) ImageNet
17. What is the full form of LFW?
(A) Local Face Window
(B) Labeled Faces in the Wild
(C) Large Face Width
(D) Low-Frequency Wavelet
18. In face detection, what does the term “sliding window” refer to?
(A) A UI component
(B) A method for thresholding
(C) A region scanned over the image to detect faces
(D) A smoothing filter
19. What is the purpose of face alignment in recognition systems?
(A) Resize images
(B) Enhance brightness
(C) Normalize facial landmarks
(D) Convert to grayscale
20. Which algorithm transforms face images into a lower-dimensional space using class labels?
(A) PCA
(B) LDA
(C) KNN
(D) HOG
21. Which feature descriptor divides the image into small cells and calculates the gradient histogram?
(A) Haar
(B) LBP
(C) HOG
(D) SIFT
22. Which of the following can be used for real-time face detection?
(A) Viola-Jones
(B) PCA
(C) HOG
(D) DeepFace
23. What is the role of non-max suppression in face detection?
(A) Enhance image edges
(B) Detect multiple face classes
(C) Eliminate redundant overlapping bounding boxes
(D) Normalize pixel values
24. Which deep learning model maps face images to a compact Euclidean space?
(A) VGGFace
(B) Eigenfaces
(C) FaceNet
(D) HOG
25. Which preprocessing step is essential for improving face recognition accuracy?
(A) Rotation
(B) Face alignment
(C) Binarization
(D) Upsampling
26. Which method converts an image into a binary pattern to describe its texture?
(A) LBP
(B) HOG
(C) PCA
(D) CNN
27. Which system uses 128-dimensional face embeddings for recognition?
(A) Eigenfaces
(B) HOG
(C) FaceNet
(D) LBP
28. What is the main limitation of traditional face recognition techniques?
(A) High computational cost
(B) Sensitivity to lighting, pose, and occlusion
(C) Inability to train
(D) Low memory usage
29. Which method relies on convolutional layers to extract hierarchical features from face images?
(A) PCA
(B) CNN
(C) KNN
(D) Gabor Filters
30. Which technique enhances the speed of detection in Viola-Jones algorithm?
(A) Integral Image
(B) Gradient Descent
(C) Convolution
(D) Fourier Transform
31. In which phase is a classifier trained on positive and negative samples in face detection?
(A) Preprocessing
(B) Training
(C) Feature extraction
(D) Testing
32. What is the goal of face verification?
(A) Detect face in image
(B) Match a face against a known identity
(C) Classify face emotions
(D) Cluster face images
33. What is the function of the softmax layer in a face recognition CNN?
(A) Normalize the input image
(B) Extract features
(C) Output class probabilities
(D) Detect edges
34. Which term refers to identifying a person in a group of faces?
(A) Face detection
(B) Face clustering
(C) Face recognition
(D) Face verification
35. What is the role of data augmentation in face recognition?
(A) Compress images
(B) Add noise to training data
(C) Improve generalization by increasing diversity
(D) Reduce training time
36. Which face descriptor technique is rotation invariant?
(A) Haar
(B) HOG
(C) LBP
(D) PCA
37. Which one is a face anti-spoofing technique?
(A) Viola-Jones
(B) Depth analysis
(C) HOG
(D) Eigenfaces
38. Which step helps reduce the dimensionality of face data?
(A) LDA
(B) FaceNet
(C) CNN
(D) Softmax
39. Which of the following techniques is least affected by partial occlusion?
(A) PCA
(B) CNN-based methods
(C) Eigenfaces
(D) LBP
40. Which face recognition method is more interpretable and computationally simpler?
(A) LBP
(B) CNN
(C) DeepFace
(D) FaceNet
41. What is the main disadvantage of Eigenface method?
(A) High training time
(B) Sensitive to illumination and pose
(C) Complex architecture
(D) No feature reduction
42. Which component in CNN learns to detect features like eyes, nose, and mouth?
(A) Activation layer
(B) Convolutional layer
(C) Fully connected layer
(D) Pooling layer
43. Which loss function is used in FaceNet to improve discrimination between faces?
(A) Binary Crossentropy
(B) Triplet Loss
(C) Hinge Loss
(D) MSE Loss
44. Which software library provides face recognition using dlib and face_recognition API?
(A) OpenCV
(B) TensorFlow
(C) Dlib
(D) face_recognition
45. Which system identifies a person across multiple surveillance cameras?
(A) Face detection
(B) Face clustering
(C) Face verification
(D) Face re-identification
46. Which process assigns identities to unknown face data?
(A) Clustering
(B) Verification
(C) Classification
(D) Detection
47. Which type of facial recognition method uses unlabeled data?
(A) Supervised learning
(B) Semi-supervised learning
(C) Unsupervised learning
(D) Reinforcement learning
48. Which factor significantly influences face recognition performance in real-world applications?
(A) Dataset size
(B) Image resolution
(C) Environmental conditions
(D) Frame rate
49. Which layer in CNN helps reduce spatial dimensions of feature maps?
(A) Convolution
(B) Dropout
(C) Pooling
(D) Dense
50. Which CNN-based model is widely used in mobile devices for efficient face recognition?
(A) VGG
(B) ResNet
(C) MobileNet
(D) AlexNet
