Abstract: Apparatuses, systems, and techniques are presented to generate digital content. In at least one embodiment, one or more neural networks are used to generate video information based at least in part upon voice information and a combination of image features and facial landmarks corresponding to one or more images of a person.

Abstract: A method of performing air pollution estimation is provided. The method is to be implemented using a processor of a computer device and includes: generating a spectral image based on an original color image of an environment under test using a spectral transformation matrix; supplying the spectral image as an input into an estimating model for air pollution estimation; and obtaining an estimation result from the estimating model indicating a degree of air pollution of the environment under test.

Abstract: A microfluidic detection unit comprises at least one fluid injection section, a fluid storage section and a detection section. Each fluid injection section defines a fluid outlet; the fluid storage section is in gas communication with the atmosphere and defines a fluid inlet; the detection section defines a first end in communication with the fluid outlet and a second end in communication with the fluid inlet. A height difference is defined between the fluid outlet and the fluid inlet along the direction of gravity. When a first fluid is injected from the at least one fluid injection section, the first fluid is driven by gravity to pass through the detection section and accumulate to form a droplet at the fluid inlet, such that a state of fluid pressure equilibrium of the first fluid is established.

Abstract: Apparatuses, systems, and techniques to generate a video using two or more images comprising objects to be included in the video. In at least one embodiment, objects are identified in two or more images using one or more neural networks, to generate a video to include the objects in the video.

Abstract: Apparatuses, systems, and techniques are presented to synthesize consistent images or video. In at least one embodiment, one or more neural networks are used to generate one or more second images based, at least in part, on one or more point cloud representations of one or more first images.

Abstract: A semiconductor device includes a transistor having a source/drain and a gate. The semiconductor device also includes a conductive contact for the transistor. The conductive contact provides electrical connectivity to the source/drain or the gate of the transistor. The conductive contact includes a plurality of barrier layers. The barrier layers have different depths from one another.

Abstract: A semiconductor device includes a transistor having a source/drain and a gate. The semiconductor device also includes a conductive contact for the transistor. The conductive contact provides electrical connectivity to the source/drain or the gate of the transistor. The conductive contact includes a plurality of barrier layers. The barrier layers have different depths from one another.

Abstract: A detection system and method for the migrating cell is provided. The system is configured to detect a migrating cell combined with an immunomagnetic bead. The system includes a platform, a microchannel, a magnetic field source, a coherent light source and an optical sensing module. The microchannel is configured to allow the migrating cell to flow in it along a flow direction. The magnetic field source is configured to provide magnetic force to the migrating cell combined with the immunomagnetic bead. The magnetic force includes at least one magnetic force component and the magnetic force component is opposite to the flow direction of the microchannel. The coherent light source is configured to provide the microchannel with the coherent light. The optical sensing module is configured to receive the interference light caused by the coherent light being reflected by the sample inside the microchannel.

Abstract: A digital-to-analog converter and an operation method thereof are provided. The digital-to-analog converter includes a current source module, a decoder, a change indicator, and a random number generator. The decoder is coupled to the current source module and receives a digital input signal. The change indicator is coupled to the decoder and provides an indication signal to the decoder. The random number generator is coupled to the change indicator and provides a random number signal to the change indicator. The change indicator generates an indication signal according to the random number signal, and the decoder generates a control signal to the current source module according to the digital input signal and the indication signal, so that the current source module generates an analog output signal corresponding to the digital input signal according to the control signal.

Abstract: Apparatuses, systems, and techniques are presented to generate image data. In at least one embodiment, one or more neural networks are used to cause a lighting effect to be applied to one or more objects within one or more images based, at least in part, on synthetically generated images of the one or more objects.

Abstract: A semiconductor device structure is provided. The semiconductor device structure includes a contact layer over a metal silicide layer. The contact layer extends through a first dielectric structure. The semiconductor device structure includes a first metal nitride barrier layer over sidewalls of the contact layer. The first metal nitride barrier layer is directly adjacent to the first dielectric structure. The semiconductor device structure includes a second metal nitride barrier layer partially between the contact layer and the metal silicide layer and partially between the contact layer and the first metal nitride barrier layer. The metal silicide layer is below the first metal nitride barrier layer and the second metal nitride barrier layer.

Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.

Abstract: Apparatuses, systems, and techniques are presented to generate one or more images. In at least one embodiment, one or more neural networks are used to generate one or more images of one or more objects based, at least in part, on a model of the one or more objects and texture information.

Abstract: A semiconductor device structure is provided. The semiconductor device structure includes a dielectric structure over the substrate. The semiconductor device structure includes a contact structure passing through the dielectric structure. The contact structure includes a contact layer, a first barrier layer, and a second barrier layer, the contact layer passes through the first barrier layer, the first barrier layer passes through the second barrier layer, the first barrier layer surrounds the contact layer, the second barrier layer surrounds a first upper portion of a sidewall of the first barrier layer and exposes a first lower portion of the sidewall of the first barrier layer, and the sidewall faces away from the contact layer.

Abstract: Apparatuses, systems, and techniques for texture synthesis from small input textures in images using convolutional neural networks. In at least one embodiment, one or more convolutional layers are used in conjunction with one or more transposed convolution operations to generate a large textured output image from a small input textured image while preserving global features and texture, according to various novel techniques described herein.

Abstract: A semiconductor device structure is provided. The semiconductor device structure includes a dielectric structure over the substrate. The semiconductor device structure includes a contact structure passing through the dielectric structure. The contact structure includes a contact layer, a first barrier layer, and a second barrier layer. The first barrier layer surrounds the contact layer, the second barrier layer surrounds a first upper portion of the first barrier layer, the contact layer passes through the first barrier layer and extends into the dielectric structure, and the first barrier layer passes through the second barrier layer and extends into the dielectric structure.

Abstract: Apparatuses, systems, and techniques are presented to reconstruct one or more images. In at least one embodiment, one or more neural networks are used to generate one or more images of one or more objects based, at least in part, on input indicating motion of the one or more objects.

Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.

Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.

Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.