Ðн¡½²Ì³Ñ§Êõ½²×ù

¹¦·ò:   2018Äê11ÔÂ7ÈÕ£¨ÖÜÈý£©ÏÂÎç14:00-17:00

µØÖ·:   У±¾²¿¶«ÇøÏèӢ¥T706ÊÒ

½²×ù:   »ùÓÚ¸ÐÖªÖÊÁ¿µÄÊÓƵ±àÂëÓÅ»¯

Ñݽ²Õß:   ÃÀ¹úÄϼÓÖÝ´óѧµç×Ó¹¤³Ìϵ Ñ ²©Ê¿

Ñݽ²Õß¼ò½é£ºChao Yang received his B.E. degree in Communication Engineering and Ph. D. degree in Communication and Information Systems from Shanghai University, Shanghai, China, in 2012 and 2017 respectively. From Nov. 2017 to Oct. 2018, he was a postdoctoral researcher at Media Communications Lab (MCL) in University of Southern California (USC), Los Angeles, USA. His research interests include video compression and video processing.

½²×ùÌáÒª£ºVideo compression has becoming more and more important due to the increasing demand for high-quality video services. The traditional video coding technologies like h.264/ h.265 have made great progress in video compression. Yet these video coding technologies mainly focus on compression video while achieving high objective quality, like PSNR. In this talk, I will introduce a data-driven Just-Noticeable-Difference (JND) based video coding, which mainly focuses on the subjective quality of compressed video. based on Satisfied User Ratio (SUR), video stream bitrate can be greatly reduced without perceptual loss.

Ô¼ÇëÈË£ºÍ¨Ñ¶ÓëÐÅÏ¢¹¤³ÌѧԺ  ÉòÀñȨ×êÑÐÔ±

Ó­½Ó¿í´óÀÏʦºÍѧÉú²ÎÓ룡