US 10,890,966 B2
Graphics processing systems
Daren Croxford, Swaffham Prior (GB)
Assigned to Arm Limited, Cambridge (GB)
Filed by ARM Limited, Cambridge (GB)
Filed on Jul. 27, 2017, as Appl. No. 15/661,213.
Prior Publication US 2019/0033961 A1, Jan. 31, 2019
Int. Cl. G06F 3/01 (2006.01); G06T 5/00 (2006.01); A63F 13/53 (2014.01); G06T 1/20 (2006.01); G06T 3/20 (2006.01); G06T 7/30 (2017.01); G06T 7/246 (2017.01); A63F 13/25 (2014.01); G02B 27/01 (2006.01); A63F 13/211 (2014.01); A63F 13/212 (2014.01); A63F 13/5255 (2014.01)
CPC G06F 3/012 (2013.01) [A63F 13/211 (2014.09); A63F 13/212 (2014.09); A63F 13/25 (2014.09); A63F 13/5255 (2014.09); A63F 13/53 (2014.09); G02B 27/017 (2013.01); G06T 1/20 (2013.01); G06T 3/20 (2013.01); G06T 7/248 (2017.01); G06T 7/30 (2017.01); G02B 2027/014 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method of operating a graphics processing system that renders a sequence of frames each representing a view of a scene of one or more objects, and generates extrapolated frames by extrapolating object motion from rendered frames, the method comprising:
generating a sequence of frames each representing a view of the same scene, wherein each frame in the sequence of frames represents a view of the scene at a respective point in time and based on a view orientation indicated by view orientation data received by the graphics processing system, and wherein the sequence of frames includes a first frame representing a view of the scene at a first point in time based on a view orientation indicated by first view orientation data received by the graphics processing system, and another frame representing a view of the scene at a second point in time based on a view orientation indicated by second view orientation data received by the graphics processing system;
generating, for each region of a set of plural regions that the first frame has been divided into, a signature indicative of and derived from the content of the region of the first frame;
generating, for each region of a set of plural regions that the other frame has been divided into, a signature indicative of and derived from the content of the region of the other frame; and
generating an extrapolated frame that represents a view of the scene at a third point in time after first and second points in time;
wherein generating the extrapolated frame comprises:
for each of at least one region of the set of plural regions that the first frame has been divided into:
comparing the signature indicative of and derived from the content of the region of the first frame with the signature indicative of and derived from the content of a corresponding region of the other frame;
determining whether the signature indicative of and derived from the content of the region of the first frame is dissimilar to the signature indicative of and derived from the content of the corresponding region of the other frame using the comparison of the signature indicative of and derived from the content of the region of the first frame with the signature indicative of and derived from the content of the corresponding region of the other frame; and
generating a region of the extrapolated frame by:
when it is determined that the signature indicative of and derived from the content of the region of the first frame is dissimilar to the signature indicative of and derived from the content of the corresponding region of the other frame using the comparison of the signature indicative of and derived from the content of the region of the first frame with the signature indicative of and derived from the content of the corresponding region of the other frame:
determining whether the region of the first frame contains one or more objects, the motion of which needs to be extrapolated to generate the extrapolated frame; and
when it is determined that the region of the first frame contains one or more objects, the motion of which needs to be extrapolated to generate the extrapolated frame:
extrapolating the motion of the one or more objects contained within the region of the first frame; and
generating the region of the extrapolated frame using the extrapolation of the motion of the one or more objects contained within the region of the first frame; and
when it is not determined that the signature indicative of and derived from the content of the region of the first frame is dissimilar to the signature indicative of and derived from the content of the corresponding region of the other frame using the comparison of the signature indicative of and derived from the content of the region of the first frame with the signature indicative of and derived from the content of the corresponding region of the other frame:
omitting determining whether the region of the first frame contains one or more objects, the motion of which needs to be extrapolated to generate the extrapolated frame;
omitting extrapolating the motion of any objects contained within the region of the first frame; and
generating the region of the extrapolated frame without using extrapolation of motion of objects contained within the region of the first frame;
the method further comprising:
generating a transformed extrapolated frame representing a view of the scene at the third point in time after the first and second points in time based on a view orientation indicated by third view orientation data received by the graphics processing system by transforming the extrapolated frame based on the view orientation indicated by the third view orientation data received by the graphics processing system; and
displaying the transformed extrapolated frame.