180 degree images1-bit matrix completion108 Megapixel1/f noise100 Hue test1st and 2nd FM generation 3D halftoning1 MICRON PIXELS133-MEGAPIXEL
2
2.5 D printing2x2 On-chip Lens2D/3D imaging, high performance computing, imaging systems, efficient computations and storage2.5D PRINTING2.5D printing2D AND 3D CONVERTIBLE DISPLAY2D2.5D2AFC2-D barcodes2D-TO-3D CONVERSION ARTIFACTS2D printing2.5D reconstruction2D VIEWS2D DCT2D to Hologram conversion2D metrics2-d scale2D and 3D video2D-plus-depth video
3
3D Gaussian splatting360-degree3D Display3D MESHES3D displays360-degree video3D models3D communications3DViewers3DSR3D-CNN3D Scene Reconstruction3D Lidar360° STEREO PANORAMAS3D Iterative Halftoning3D MESH3D reconstruction3D Compression and Encryption3D VISUALIZATION3DMM3D print3D halftoning3D-human body detection3D recursive search3-T pixel3D Video3D Meshes3D SALIENCY3D SHAPE INDEXING AND RETRIEVAL3D SCENE RECONSTRUCTION AND MODELING3D and 2D3d localization3D MODELLING3-D RECONSTRUCTION3D RANGE IMAGING3D Image Processing3D Video Conferencing3D Quality3D shape analysis360-DEGREE IMAGE360-deg quality assessment3D object shape3D Compression3D Computer Graphics3D Models3D projector35MM FILM DIGITIZATION3D mesh simplification3D human-centered technologies3D point cloud3D Scene Reconstruction and Modeling3DCNN3D-Anisotropic smoothing3D-assisted features3D localization3D/4D SCANNING3D range geometry3D digitization and dissemination3D localization and mapping3D colour Digital Image Correlation3D Print Appearance3D Photogrammetry3D RECOVERY3D printer3D INTERACTION360° VIDEO3D ACQUISITION ARCHITECTURE3D imaging3D audio3D DISPLAY3D-high efficiency video coding3D visual representation360 Video3A ALGORITHMS3-D SHAPE RECOVERY360x3D scene capture3d mapping3D PROFILE360-Degree Video Technology3D cinema and TV3D TV3D USER INTERFACES3D-printing3D Range Data Encoding3d3D CG Image360-degree imaging3D Color Printing3D Reconstruction3D COMPRESSION AND ENCRYPTION3D glasses3D surface structure based halftoning3D vision3DGS3D3D encoding3D IMAGE3d video3A Algorithms3D STIMULI360-degree Image3D model3D compression3D Imaging3D TRANSFORMATION3D Modeling3D STACK3D surface reproduction3D Range Data3D Tracking3D EDUCATIONAL MATERIAL3D Video Communications3D/4D DATA PROCESSING AND FILTERING3D Shape Indexing and Retrieval360-degree videos3D scene flow estimation3D connected tube model3D-HEVC3D adaptive halftoning3D Halftoning3D Data Sources3D camera3D Digital Image Correlation3D image compression3D shape3D PRINTER3D Communications360-degree content3D printing3D Point Cloud3D affine transformation3D recovery3D Printing3D scanning3D Mapping3D capture3D modelling360 degree images3D Range Data Compression3D digital halftoning3D Immersion3D Vision3D rigid transformations3D RECONSTRUCTION3D modeling3D skeletal joint point cloud; deep learning; PointNet++ method3d scanning3D/2D Visuals3D depth sensing3D range scanning3ARRI footage3D Curvelet360° video360 IMAGING3D HALFTONING3D refinement3D display3D-color perception3D mesh3D shape indexing and retrieval3D Data Processing3D Telepresence3D image3D DIGITIZATION METHOD FOR OIL PAINTINGS3DLUT3D position measurement of people3D-shooting3D Measurement3D theater program listing3D PRINTING3D face alignment3D MODEL3D Object Detection360-degree images3D Morphable Model3D stereo vision360 panorama3D perception3D CAMERAS360 degrees video360-degree art exhibition360-degree image projection3D-LUT3D video3D/4D Scanning360VR3D/4D Data Processing and Filtering3D mapping and localization3D depth-map360-degree video streaming360-video3D data processing3D optical scans3D Gaussian Splatting3D Saliency3D surface3D warping3D objects3D scene classification3D video processing
Our central goal was to create automatic methods for semantic segmentation of human figures in images of fine art paintings. This is a difficult problem because the visual properties and statistics of artwork differ markedly from the natural photographs widely used in research in automatic segmentation. We used a deep neural network to transfer artistic style from paintings across several centuries to modern natural photographs in order to create a large data set of surrogate art images. We then used this data set to train a separate deep network for semantic image segmentation of genuine art images. Such data augmentation led to great improvement in the segmentation of difficult genuine artworks, revealed both qualitatively and quantitatively. Our unique technique of creating surrogate artworks should find wide use in many tasks in the growing field of computational analysis of fine art.