1 MICRON PIXELS133-MEGAPIXEL1st and 2nd FM generation 3D halftoning108 Megapixel1/f noise100 Hue test1-bit matrix completion180 degree images
2
2D metrics2-d scale2D to Hologram conversion2D-plus-depth video2D and 3D video2D DCT2D VIEWS2.5D reconstruction2AFC2D2.5D2D printing2-D barcodes2D-TO-3D CONVERSION ARTIFACTS2x2 On-chip Lens2D/3D imaging, high performance computing, imaging systems, efficient computations and storage2.5D PRINTING2.5 D printing2D AND 3D CONVERTIBLE DISPLAY2.5D printing
3
3D face alignment3D PRINTING3D MODEL3D Object Detection360-degree images3D Morphable Model360 panorama3D perception3D stereo vision360-degree image projection3D CAMERAS360 degrees video360-degree art exhibition3D video3D-LUT3D/4D Scanning3D refinement3D-color perception3D display3D mesh3D shape indexing and retrieval3D Data Processing3D image3D Telepresence3D position measurement of people3D-shooting3DLUT3D DIGITIZATION METHOD FOR OIL PAINTINGS3D Measurement3D theater program listing3D data processing3D optical scans3D Saliency3D surface3D Gaussian Splatting3D objects3D warping3D scene classification3D video processing3D/4D Data Processing and Filtering360VR3D mapping and localization360-video360-degree video streaming3D depth-map3D Tracking3D EDUCATIONAL MATERIAL3D Shape Indexing and Retrieval3D/4D DATA PROCESSING AND FILTERING3D Video Communications360-degree videos3D connected tube model3D scene flow estimation3D adaptive halftoning3D-HEVC3D camera3D Halftoning3D Data Sources3D Digital Image Correlation3D image compression360-degree Image3D compression3D Imaging3D model3D TRANSFORMATION3D Modeling3D STACK3D surface reproduction3D Range Data3D digital halftoning3D Vision3D Immersion3D rigid transformations3D RECONSTRUCTION3D modeling3D skeletal joint point cloud; deep learning; PointNet++ method3D/2D Visuals3d scanning3D depth sensing3D Curvelet3ARRI footage360° video3D range scanning3D HALFTONING360 IMAGING3D PRINTER3D shape360-degree content3D Communications3D printing3D affine transformation3D Point Cloud3D Printing3D recovery3D modelling3D scanning3D Mapping3D capture3D Range Data Compression360 degree images3D range geometry3D digitization and dissemination3D localization and mapping3D colour Digital Image Correlation3D Print Appearance3D printer3D RECOVERY3D Photogrammetry3D ACQUISITION ARCHITECTURE3D audio3D imaging360° VIDEO3D INTERACTION3D mesh simplification35MM FILM DIGITIZATION3D projector3D point cloud3D human-centered technologies3DCNN3D-Anisotropic smoothing3D Scene Reconstruction and Modeling3D-assisted features3D localization3D/4D SCANNING3D Range Data Encoding3d3D-printing3D Color Printing360-degree imaging3D CG Image3D Reconstruction3D glasses3D COMPRESSION AND ENCRYPTION3D surface structure based halftoning3D encoding3D3D vision3DGS3D IMAGE3D STIMULI3d video3A Algorithms3D DISPLAY3D visual representation3D-high efficiency video coding3A ALGORITHMS360 Video3d mapping360x3D scene capture3-D SHAPE RECOVERY3D cinema and TV3D PROFILE360-Degree Video Technology3D TV3D USER INTERFACES3D-CNN3D Lidar3D Scene Reconstruction3D MESH360° STEREO PANORAMAS3D Iterative Halftoning3D reconstruction3D VISUALIZATION3D Compression and Encryption3DMM3D Gaussian splatting3D MESHES3D Display360-degree3D displays3DViewers360-degree video3D communications3D models3DSR3D MODELLING3-D RECONSTRUCTION3d localization3D RANGE IMAGING3D Image Processing3D Quality3D Video Conferencing360-DEGREE IMAGE360-deg quality assessment3D shape analysis3D object shape3D Computer Graphics3D Compression3D Models3D halftoning3D print3-T pixel3D Video3D recursive search3D-human body detection3D SALIENCY3D Meshes3D SHAPE INDEXING AND RETRIEVAL3D SCENE RECONSTRUCTION AND MODELING3D and 2D
Although considerable progress has been made in recognizing multi-character text from images, there are still cases where there is a lack of robust computationally-efficient methods that can execute on portable devices to read device displays in the wild. We specifically address the problem of parsing digits from 7 segment displays. Recognizing these displays is important for many tasks such as assisting users with tasks using augmented reality agents that need to verify actions or connecting legacy devices to the internet for process control using cheap cameras. Legacy techniques based on image processing operators and OCR are brittle whereas massive deep networks are too computationally expensive. We describe a computationally tractable VGG style backbone combined with a novel digit inference head that can be trained using a synthetic display generator with novel augmentations. We show the model trained on augmented synthetic data generalizes well to a corpus of real-world display images getting 97.8% single-frame accuracy and obtaining a throughput of 30 frames per second. We describe how the output can be further stabilized to improve accuracy through a kind of mode filtering.