Abstract: Embodiments are generally directed to an adaptive deformable kernel prediction network for image de-noising. An embodiment of a method for de-noising an image by a convolutional neural network implemented on a compute engine, the image including a plurality of pixels, the method comprising: for each of the plurality of pixels of the image, generating a convolutional kernel having a plurality of kernel values for the pixel; generating a plurality of offsets for the pixel respectively corresponding to the plurality of kernel values, each of the plurality of offsets to indicate a deviation from a pixel position of the pixel; determining a plurality of deviated pixel positions based on the pixel position of the pixel and the plurality of offsets; and filtering the pixel with the convolutional kernel and pixel values of the plurality of deviated pixel positions to obtain a de-noised pixel.

Abstract: A substrate and a display device are provided. A first film layer of the substrate is formed on a third film layer and patterned to form a target pattern. A second film layer is formed on the third film layer and the first film layer. Within a distribution area of the target pattern, film edges of the first film layer are formed with protrusion structures facing the second film layer. Based on the protrusion structures, a deformable margin of the first film layer and the second film layer at a stacking position is increased, and a pattern of the second film layer is thus prevented from being cracked to generate cracks.

Abstract: An electrode structure is disclosed, which includes a buffer layer disposed on a substrate; and an electrode disposed on a surface of the buffer layer away from the substrate, an edge of the electrode including an extension surface extending from a surface of the electrode away from the substrate, and the extension surface is in contact with a surface of the buffer layer and forms an included angle with a surface of the buffer layer contacting the electrode. An anti-reflection layer is disposed at the edge of the electrode, the anti-reflection layer surrounds and covers the edge of the electrode, and the anti-reflection layer extends to be in contact with the buffer layer. An undercut structure is formed between an outer surface of the anti-reflection layer and the surface of the buffer layer.

Abstract: An electrode structure is disclosed, which includes a buffer layer disposed on a substrate; and an electrode disposed on a surface of the buffer layer away from the substrate, an edge of the electrode including an extension surface extending from a surface of the electrode away from the substrate, and the extension surface is in contact with a surface of the buffer layer and forms an included angle with a surface of the buffer layer contacting the electrode. An anti-reflection layer is disposed at the edge of the electrode, the anti-reflection layer surrounds and covers the edge of the electrode, and the anti-reflection layer extends to be in contact with the buffer layer. An undercut structure is formed between an outer surface of the anti-reflection layer and the surface of the buffer layer.

Abstract: A substrate and a display device are provided. A first film layer of the substrate is formed on a third film layer and patterned to form a target pattern. A second film layer is formed on the third film layer and the first film layer. Within a distribution area of the target pattern, film edges of the first film layer are formed with protrusion structures facing the second film layer. Based on the protrusion structures, a deformable margin of the first film layer and the second film layer at a stacking position is increased, and a pattern of the second film layer is thus prevented from being cracked to generate cracks.