Edge-enhanced video frame blending
Apparatuses, systems, and techniques to process image frames. In at least one embodiment, one or more edges of one or more objects within one or more images are digitally enhanced before downsampling the one or more objects.
1 . One or more processors, comprising:
circuitry to cause one or more edges of one or more objects within one or more images to be digitally enhanced, wherein the one or more images are to be digitally enhanced comprising filtering one or more pixels of the one or more images to locate at least one edge of the one or more edges and further converting the one or more filtered images into one or more luma images before down-sampling the one or more converted images.
2 . The one or more processors of claim 1 , wherein the circuitry is further to cause down-sampled one or more objects to be input into a neural network to increase a framerate of a video.
3 . The one or more processors of claim 1 , wherein the one or more images comprise a plurality of images and wherein the circuitry is further to cause a neural network to generate information to be used to blend the one or more images.
4 . The one or more processors of claim 1 , wherein the circuitry is to digitally enhance the one or more edges of the one or more objects based, at least in part, on an edge-enhancing filter.
5 . The one or more processors of claim 1 , wherein digitally enhancing the one or more edges causes at least one edge to be present in a down-sampled image when down-sampling without said digitally enhancing would cause said at least one edge to be modified by said down-sampling.
6 . The one or more processors of claim 1 , wherein the circuitry is further to cause one or more down-sampled objects to be of a different color space than the one or more images.
7 . The one or more processors of claim 1 , wherein the one or more images are motion warped images generated, based at least in part, on a set of video frames.
8 . A computer-implemented method comprising:
causing one or more edges of one or more objects within one or more images to be digitally enhanced, wherein the one or more images are to be digitally enhanced comprising filtering one or more pixels of the one or more images to locate at least one edge of the one or more edges and further converting the one or more filtered images into one or more luma images before down-sampling the one or more converted images.
9 . The computer-implemented method of claim 8 , further comprising inputting one or more down-sampled objects into one or more neural networks to generate information indicating how to blend the one or more images.
10 . The computer-implemented method of claim 8 , further comprising generating the one or more images by at least using motion information to warp one or more other images.
11 . The computer-implemented method of claim 8 , wherein the one or more images comprise two or more images generated based, at least in part, on a same image.
12 . The computer-implemented method of claim 8 , wherein the one or more images comprise two or more successive video frames from a video.
13 . The computer-implemented method of claim 8 , further comprising generating another image based, at least in part, on one or more down-sampled objects.
14 . The computer-implemented method of claim 8 , wherein down-sampling the one or more images preserves one or more edges that would not be preserved without enhancement of the one or more edges.
15 . A computer system comprising:
one or more processors and memory storing executable instructions that, if performed by the one or more processors, are to cause one or more edges of one or more objects within one or more images to be digitally enhanced, wherein the one or more images are to be digitally enhanced comprising filtering one or more pixels of the one or more images to locate at least one edge of the one or more edges and further converting the one or more filtered images into one or more luma images before down-sampling the one or more converted images.
16 . The computer system of claim 15 , wherein the instructions are further to cause the one or more down-sampled images to be used by one or more neural networks to generate information to blend the one or more images.
17 . The computer system of claim 15 , wherein the instructions are further to cause down-sampled one or more objects to increase a framerate of a video.
18 . The computer system of claim 15 , wherein the instructions are further to cause the one or more images to be interpolated based, at least in part, on down-sampled one or more objects.
19 . The computer system of claim 15 , wherein down-sampled one or more objects are within the one or more down-sampled images.
20 . The computer system of claim 15 , wherein the one or more images are motion-warped images generated from a same image.