Abstract There are billions of photographs on the Internet, comprising the largest and most diverse photo collection ever assembled. How can computer vision researchers exploit thi...
Scalable image retrieval systems usually involve hierarchical quantization of local image descriptors, which produces a visual vocabulary for inverted indexing of images. Although ...
Rongrong Ji (Harbin Institute of Technology), Xing...
Abstract—We have developed a new face hallucination framework termed from local pixel structure to global image super-resolution (LPS-GIS). Based on the assumption that two simil...
We study in this paper the problem of bridging the semantic gap between low-level image features and high-level semantic concepts, which is the key hindrance in content-based imag...
Abstract. We present a first system to semi-automatically create a visual representation for a given, short text. We first parse the input text, decompose it into suitable units,...
Katharina Schwarz, Pavel Rojtberg, Joachim Caspar,...