Human-nameable visual “attributes” can benefit various recognition tasks. However, existing techniques restrict these properties to categorical labels (for example, a person ...
Traditional supervised visual learning simply asks annotators “what” label an image should have. We propose an approach for image classification problems requiring subjective...
ABSTRACT: OCR is an error-prone process. It is time-consuming and expensive to manually proofread OCR results. The errors remaining in OCRed texts can cause serious problems in rea...
To have a robust and informative image content representation for image categorization, we often need to extract as many as possible visual features at various locations, scales a...
There is a lack of an integrated technology that will increase effective usage of the vast and heterogeneous multi-lingual and multimedia digital content. The need is being express...