yoloYouTubeYou only look once v3 (YOLOv3)YOLOv5YOLOv8YOLO-RYOLOYolov5You only look once (YOLO)YOLOX
Z
Z numberZEISSZERNIKEZoom FatigueZero Parallax SettingZ-type SchlierenZero-Shot ClassificationZERNIKE MOMENTSZernike polynomialZebra Embryo
0
0.8um pixel
1
180 degree images1-bit matrix completion100 Hue test1/f noise108 Megapixel1st and 2nd FM generation 3D halftoning133-MEGAPIXEL1 MICRON PIXELS
2
2.5D printing2D AND 3D CONVERTIBLE DISPLAY2.5 D printing2.5D PRINTING2D/3D imaging, high performance computing, imaging systems, efficient computations and storage2D-TO-3D CONVERSION ARTIFACTS2-D barcodes2D printing2.5D2D2AFC2.5D reconstruction2D VIEWS2D DCT2D and 3D video2D-plus-depth video2D to Hologram conversion2-d scale2D metrics
3
3D and 2D3D SCENE RECONSTRUCTION AND MODELING3D SHAPE INDEXING AND RETRIEVAL3D SALIENCY3D Meshes3D-human body detection3D recursive search3D Video3-T pixel3D print3D halftoning3D Models3D Compression3D Computer Graphics3D object shape3D shape analysis360-deg quality assessment360-DEGREE IMAGE3D Video Conferencing3D Quality3D Image Processing3D RANGE IMAGING3d localization3-D RECONSTRUCTION3D MODELLING3DSR3D communications360-degree video3D displays360-degree3D MESHES3D Display3D Gaussian splatting3DMM3D Compression and Encryption3D VISUALIZATION3D reconstruction3D Iterative Halftoning360° STEREO PANORAMAS3D MESH3D Lidar3D Scene Reconstruction3D-CNN3D USER INTERFACES3D TV360-Degree Video Technology3D PROFILE3D cinema and TV360x3D scene capture3-D SHAPE RECOVERY3d mapping360 Video3A ALGORITHMS3D-high efficiency video coding3D visual representation3D DISPLAY3d video3D STIMULI3D IMAGE3D vision3D3D encoding3D surface structure based halftoning3D glasses3D COMPRESSION AND ENCRYPTION3D Reconstruction360-degree imaging3D Color Printing3D-printing3d3D Range Data Encoding3D/4D SCANNING3D localization3D-assisted features3D Scene Reconstruction and Modeling3D-Anisotropic smoothing3DCNN3D point cloud3D human-centered technologies3D projector3D mesh simplification35MM FILM DIGITIZATION3D INTERACTION360° VIDEO3D audio3D imaging3D ACQUISITION ARCHITECTURE3D RECOVERY3D printer3D Print Appearance3D colour Digital Image Correlation3D localization and mapping3D digitization and dissemination3D range geometry360 degree images3D Range Data Compression3D capture3D Mapping3D scanning3D modelling3D recovery3D Printing3D Point Cloud3D affine transformation3D printing3D Communications360-degree content3D shape3D PRINTER360 IMAGING3D HALFTONING3D range scanning360° video3D Curvelet3ARRI footage3D depth sensing3D/2D Visuals3D modeling3D RECONSTRUCTION3D rigid transformations3D Vision3D Immersion3D digital halftoning3D Range Data3D surface reproduction3D STACK3D Modeling3D TRANSFORMATION3D model3D Imaging3D compression360-degree Image3D image compression3D Digital Image Correlation3D Data Sources3D Halftoning3D camera3D-HEVC3D adaptive halftoning3D scene flow estimation3D connected tube model360-degree videos3D Video Communications3D Shape Indexing and Retrieval3D/4D DATA PROCESSING AND FILTERING3D EDUCATIONAL MATERIAL3D Tracking360-degree video streaming3D depth-map360-video3D mapping and localization3D/4D Data Processing and Filtering360VR3D video processing3D scene classification3D warping3D objects3D surface3D Saliency3D optical scans3D data processing3D theater program listing3D Measurement3D DIGITIZATION METHOD FOR OIL PAINTINGS3D-shooting3D position measurement of people3D Telepresence3D Data Processing3D shape indexing and retrieval3D mesh3D-color perception3D display3D refinement3D/4D Scanning3D video3D-LUT360 degrees video360-degree art exhibition3D CAMERAS360-degree image projection3D stereo vision3D perception360 panorama3D Morphable Model360-degree images3D Object Detection3D MODEL3D PRINTING3D face alignment
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.