A Low-Complexity Screen Compression Scheme for Interactive Screen Sharing

Subjective quality evaluation of compressed digital compound images

“…We employ five compression methods to generate distorted images: JPEG, JPEG2000, H.264 intra coding, DjVu [13,42,43] and LSC [4]. The JPEG, JPEG2000 and H.264 intra coding are standard algorithms to encode natural images.…”

“…For efficient transmission among different devices, transmission systems should be deployed in a platform-free manners (without consideration of different protocols of different systems) and DCIs have to be highly compressed. For this requirement, cloud computing systems provide a strategy to share DCIs from the cloud to thin-clients; such systems render visual contents as DCIs in the cloud and then transmit them after compression to clients [4]. Many other computer generated compound images, such as promotional posters and microblogs, are also represented and stored in the form of DCIs.…”

“…To address this problem, many studies [4,5,6,7,8] propose segmentation techniques to decompose DCIs into textual and of high consistency with human visual perception when viewing distorted natural images, such as SSIM [27], VIF [28], FSIM [29], MAD [30], GSIM [31] and BLIINDS-II [32].…”

“…Different characteristics of textual and pictorial regions result in different visual experience for human beings, and this requires different compression methods to the two parts separately. Some algorithms are proposed to segment and compress the textual and pictorial parts separately for DCIs [4,5,6,7,8]. However, there is no study investigating how to compare the compression performance for DCIs with respect to the characteristics of human visual perception.…”

“…• LSC: In the LSC algorithm [4], a block segmentation algorithm is firstly utilized to separate textual blocks from pictorial blocks in DCIs, and then JPEG and a specialized Basic Color and Index Map (BCIM) codec [4] are used to encode the pictorial and textual blocks, respectively. Eight distorted images are generated for each reference image via the LSC algorithm.…”