1. Which technique is commonly used in image inpainting to fill missing areas using neighboring pixel information?
(A) Histogram Equalization
(B) Convolutional Filtering
(C) Patch-based Method
(D) Contrast Stretching
2. What is the primary objective of image inpainting?
(A) Image segmentation
(B) Object detection
(C) Restoration of missing or damaged regions
(D) Enhancement of image resolution
3. Which mathematical tool is frequently used in PDE-based image inpainting?
(A) Laplace Equation
(B) Fourier Transform
(C) Wavelet Transform
(D) Sobel Operator
4. Which of the following is a deep learning approach used for image inpainting?
(A) Canny Edge Detector
(B) Generative Adversarial Network
(C) Hough Transform
(D) PCA
5. What does the term “contextual attention” relate to in deep learning-based inpainting methods?
(A) Enhancing resolution
(B) Focusing on important image boundaries
(C) Matching similar features from non-masked regions
(D) Generating histogram maps
6. Which approach utilizes known pixels surrounding the target area to iteratively reconstruct missing regions?
(A) Frequency Domain Filtering
(B) Exemplar-based Inpainting
(C) Optical Flow Estimation
(D) Principal Component Analysis
7. Which type of neural network is most suitable for learning semantic context in image inpainting tasks?
(A) RNN
(B) CNN
(C) SVM
(D) KNN
8. Which loss function is commonly used in GAN-based inpainting methods to improve realism?
(A) L1 Loss
(B) Binary Cross Entropy
(C) Adversarial Loss
(D) Mean Squared Error
9. Which of the following can be considered a major challenge in image inpainting?
(A) Memory allocation
(B) Color saturation
(C) Structural consistency in filled areas
(D) Bit depth limitation
10. Which metric is typically used to evaluate the performance of image inpainting?
(A) PSNR
(B) SSIM
(C) Both A and B
(D) MAE only
11. Which algorithm is widely used in traditional exemplar-based inpainting?
(A) PatchMatch
(B) K-means
(C) SIFT
(D) Backpropagation
12. Which region is targeted for modification during the image inpainting process?
(A) Histogram bins
(B) Boundary edges
(C) Masked or missing area
(D) Color space
13. Which method uses diffusion processes to propagate information from known regions to unknown ones?
(A) Texture synthesis
(B) PDE-based Inpainting
(C) Edge detection
(D) Clustering
14. What role does the discriminator play in GAN-based inpainting models?
(A) Fills the missing parts
(B) Determines image resolution
(C) Differentiates between real and inpainted regions
(D) Compresses image data
15. Which aspect of image inpainting ensures that the filled region blends well with the surrounding texture?
(A) Histogram Matching
(B) Edge-preserving Filtering
(C) Texture Coherence
(D) Image Thresholding
16. In semantic inpainting, what guides the filling of large missing regions?
(A) RGB color patterns
(B) Low-level features only
(C) High-level semantic information
(D) Binary mask alone
17. Which method involves selecting the most similar patch from the known area to fill the missing area?
(A) Patch Prior
(B) Patch Propagation
(C) Patch Matching
(D) Patch Embedding
18. In what scenario is image inpainting particularly useful?
(A) Histogram equalization
(B) Object segmentation
(C) Removing unwanted objects from images
(D) Reducing noise
19. Which of the following models uses attention mechanisms for improved inpainting?
(A) Autoencoder
(B) CycleGAN
(C) Contextual Attention Model
(D) VGGNet
20. What does a mask represent in an inpainting task?
(A) Color enhancement map
(B) Region to be ignored
(C) Region to be filled
(D) Gaussian weight
21. Which type of inpainting technique focuses more on capturing fine texture details?
(A) Structure-preserving
(B) Texture synthesis
(C) Depth mapping
(D) Color mapping
22. What is the role of an encoder in an encoder-decoder inpainting architecture?
(A) Generate noise
(B) Extract feature representations
(C) Apply filters
(D) Measure brightness
23. Which part of the image inpainting model reconstructs the masked region using learned features?
(A) Feature extractor
(B) Encoder
(C) Decoder
(D) Classifier
24. In GAN-based inpainting, which component generates plausible content for the missing regions?
(A) Classifier
(B) Generator
(C) Discriminator
(D) Transformer
25. Which type of convolution is sometimes used in inpainting to handle irregular masks?
(A) Dilated Convolution
(B) Masked Convolution
(C) Transposed Convolution
(D) Depthwise Convolution
26. How does total variation loss help in image inpainting?
(A) Increases resolution
(B) Encourages spatial smoothness
(C) Improves color accuracy
(D) Detects edges
27. Which traditional technique works well for small missing regions in an image?
(A) Median filtering
(B) Diffusion-based inpainting
(C) Frequency domain analysis
(D) Super-resolution
28. Why are perceptual losses used in deep learning-based inpainting models?
(A) For sharpening edges
(B) To improve numerical precision
(C) To match visual quality of real images
(D) To detect object boundaries
29. Which characteristic is essential for a good inpainting result?
(A) High contrast
(B) Sharp histograms
(C) Perceptual realism
(D) Flat lighting
30. Which domain do most deep learning-based inpainting methods operate in?
(A) Frequency domain
(B) RGB space
(C) Spatial domain
(D) Time domain
31. Which technique allows iterative improvement in patch-based inpainting?
(A) Histogram Normalization
(B) PatchMatch
(C) Non-max Suppression
(D) Fast Fourier Transform
32. Which of these is a limitation of early diffusion-based inpainting methods?
(A) Over-smoothing
(B) High resolution
(C) Color distortion
(D) Histogram flattening
33. Which architecture helps capture long-range dependencies in inpainting?
(A) ResNet
(B) LSTM
(C) Transformer
(D) U-Net
34. Which type of image inpainting method requires labeled training data?
(A) Rule-based
(B) Hand-crafted
(C) Supervised learning
(D) Diffusion-based
35. Which approach improves global structure during inpainting?
(A) Texture blending
(B) Semantic mapping
(C) Global-local discriminator
(D) Histogram equalization
36. Which evaluation metric assesses structural similarity in inpainted images?
(A) MAE
(B) RMSE
(C) SSIM
(D) L2 norm
37. Which application benefits from robust image inpainting?
(A) Color segmentation
(B) Text removal in documents
(C) Edge detection
(D) Motion tracking
38. What is the function of a coarse-to-fine inpainting framework?
(A) Converts image to grayscale
(B) Builds image in multiple resolution levels
(C) Segments image into patches
(D) Applies thresholding
39. What is one drawback of GAN-based inpainting methods?
(A) High noise tolerance
(B) Limited scalability
(C) Training instability
(D) Lack of gradient computation
40. Which property helps a model generalize well to different masks during inpainting?
(A) Fixed mask shape
(B) Mask diversity
(C) Low-dimensional feature space
(D) Static learning rate
41. Which method would be best for inpainting structured content like buildings?
(A) Random noise filling
(B) Semantic segmentation-guided inpainting
(C) Contrast adjustment
(D) Histogram thresholding
42. Which stage in GAN-based inpainting adds finer details?
(A) Encoder
(B) Decoder
(C) Refinement network
(D) Downsampler
43. Which optimization method is often used to train deep inpainting models?
(A) K-means
(B) Stochastic Gradient Descent
(C) A* Search
(D) Newton-Raphson
44. What is the importance of using skip connections in encoder-decoder inpainting networks?
(A) Reduces computational time
(B) Avoids overfitting
(C) Preserves spatial details
(D) Limits mask size
45. Which dataset is commonly used for evaluating image inpainting models?
(A) MNIST
(B) COCO
(C) Pascal VOC
(D) FERET
46. What is a benefit of two-stage inpainting networks?
(A) Requires fewer layers
(B) Avoids using masks
(C) First stage focuses on structure, second on texture
(D) Uses grayscale only
47. Which filter is applied to smooth out pixel noise during preprocessing in inpainting?
(A) Mean filter
(B) Unsharp mask
(C) High-pass filter
(D) Laplacian
48. Which module in inpainting helps align features from masked and unmasked regions?
(A) Feature Attention Module
(B) Histogram Equalizer
(C) Blur Filter
(D) Edge Detector
49. What does a higher PSNR value indicate in inpainting results?
(A) Poor quality
(B) Better similarity to ground truth
(C) High frequency loss
(D) Reduced brightness
50. Which deep learning technique is often employed to improve structural coherence in image inpainting?
(A) Recurrent Neural Networks
(B) Graph Convolutional Networks
(C) Attention Mechanisms
(D) Decision Trees
