Fusing Global and Local Features for Generalized AI-Synthesized Image Detection.
Yan JuShan JiaLipeng KeHongfei XueKoki NaganoSiwei LyuPublished in: CoRR (2022)
Keyphrases
- image features
- global features
- low level
- image data
- extracting features
- multiscale
- input image
- test images
- image content
- adaboost classifier
- image regions
- global structure
- image classification
- false positives
- single image
- feature values
- image analysis
- original images
- invariant features
- image pixels
- spatial information
- feature extraction
- image representation
- keypoints
- image matching
- edge detection
- cluttered scenes
- extracted features
- image structure
- automatic image orientation
- feature representation
- artificial intelligence
- similarity measure
- partial occlusion
- image segmentation
- road signs
- detection method
- sift descriptors
- feature vectors
- feature points
- intensity histogram
- synthesized images
- feature space
- global shape
- high resolution
- feature set
- visual features
- segmentation algorithm
- hough transform
- bounding box
- image set
- feature detection
- spatial relationships