Title: "Fusion-based techniques in computer vision"
Speaker: Cosmin Ancuti
Location: "Shannon" Seminar Room, Place du Levant 3, Maxwell Building, 1st floor
Date / Time (duration): Friday 11/09/2015, 11h00 (~45')
Abstract: Image fusion is a well-studied process that aims to blend seamlessly the input information by preserving only the specific features of the composite output image. In general image fusion combines multiple-source complementary imagery in order to enhance the information apparent in the respective source images, as well as to increase the reliability of interpretation and classification.
Different than most of the existing methods we present several fusion-based image enhancing techniques that process only a single input image. In our strategies the original image is processed in several input versions that aim to enhance some specific regions. To blend effectively the information of the derived inputs in order to preserve the regions with good visibility, we filter their important features by computing several measures (weight maps). To minimize artifacts introduced by the weight maps, our approaches are designed in multi-scale fashion. We demonstrate the utility of our single image-based fusion approaches for several enhancing applications: color-to-grayscale, dehazing, underwater image/video enhancement. Moreover, our fusion-based enhancing techniques shown to improve considerably the performance of several challenging computer vision applications.
Last updated September 01, 2015, at 12:37 PM