Percorrer por autor "Santos, João M."
A mostrar 1 - 3 de 3
Resultados por página
Opções de ordenação
- Lossless coding of light field images based on minimum-rate predictorsPublication . Santos, João M.; Assunção, Pedro; Cruz, Luis A. da Silva; de Oliveira Pegado de Noronha E Távora, Luís Miguel; Fonseca-Pinto, Rui; Faria, SergioRecent developments in light field acquisition and computational photography are driving new research efforts on light field encoding methods, capable of exploiting the specific features of this type of visual data. This paper presents a research study of lossless light field image compression, using Minimum-Rate Predictors (MRP) and mainstream image and video encoders. The research is focused on three light field representation formats: lenslet images, stack of sub-aperture images and epipolar images. The main contributions of this work are the ‘Spiral-blackend’ serialization method and the use of MRP for the lossless compression of light fields with joint encoding of RGB data. The results show that the lenslet format yields lower compression efficiencies than other formats. Furthermore, it is demonstrated that the MRP algorithm consistently outperforms HEVC-RExt, JPEG2000, JPEG-LS and CALIC when light fields are represented by either a stack of sub-aperture or epipolar images.
- A method to improve HEVC lossless coding of volumetric medical imagesPublication . F.R. Guarda, André F. R.; Santos, João M.; Cruz, Luís A. da Silva; Assunção, Pedro A. A.; Rodrigues, Nuno M. M.; Faria, Sérgio M. M. deMedical imaging technology and applications are continuously evolving, dealing with images of increasing spatial and temporal resolutions, which enable more accurate processing, feature analysis and medical diagnosis. However, the increase in resolution also requires a growing amount of data to be stored, processed and exchanged or transmitted through networks. Despite the high coding efficiency achieved by the most recent image and video coding standards in lossy compression, they are not well suited for quality-critical medical image compression where either near-lossless or lossless coding is required. In this paper we propose a method to improve lossless coding of volumetric medical images, such as Magnetic Resonance and Computed Tomography, and medical sequences such as X-ray Angiography images, using the latest standard High Efficiency Video Encoder (HEVC). New pixel-wise prediction techniques are proposed to extend the current HEVC lossless tools, based on Least-Squares Prediction (LSP). Experimental results show a bitrate reduction of over 44%, when compared to DICOM recommended encoders, and 13.8% when compared to standard lossless HEVC, for 8 bpp volumetric images, and over 8% and 4.6%, respectively, for volumetric images using more than 8 bpp.
- A novel trilateral filter technique for depth map processing in 3D video codingPublication . Guarda, André F. R.; Santos, João M.; Graziosi, Danillo B.; Rodrigues, Nuno M. M.; Faria, Sérgio M. M.Free viewpoint video communication is one of the most interesting schemes for an immersive 3D experience, but it requires multiple views of the same scene. In order to limit the number of transmitted views, depth representations of the scene are used for depth image-based rendering, to generate virtual views at the receiver. Nevertheless, the video-plus-depth approach presents an additional burden of data that may be attenuated by coding down-sampled depth maps. This is able to achieve significant bit rate savings, specially for higher compression ratios, at the expense of some degradation on the quality of the decoded depth maps. However, as these techniques heavily rely on the quality of the depth maps, post-processing methods, such as filtering, are used to restore its quality. In this work we propose a new depth map processing method that is able to improve the image rendering quality. Experimental results show that a better rendering quality is achieved, when compared to other commonly used filters.