Abstract: This application relates to a data storage method, a reading method, an apparatus, a storage medium, and a program product. The data storage method is applied to a processor and includes: (S1) obtaining event data; (S2) processing the event data, to obtain a plurality of data blocks and a plurality of corresponding metadata blocks, where each data block includes a part of processed event data, and each metadata block includes a start time, an end time, a quantity, and a compression manner of event data in a data block corresponding to the metadata block, and a storage address of the data block; and (S3) storing the plurality of data blocks into a first storage area, and storing the plurality of metadata blocks into a second storage area.

Abstract: This application provides an event data processing method, including: obtaining event data shot by an event camera, where the event data includes a timestamp of occurrence of an event, pixel coordinates, and a brightness change status; and generating an event frame matrix based on a plurality of pieces of event data within a preset time period, where a location of a matrix element in the event frame matrix corresponds to pixel coordinates, a value of the matrix element indicates whether there is an event at a corresponding pixel and an event polarity of a pixel with an event, if there is an event at the pixel, it indicates that brightness of the pixel is changed within the preset time period, or if there is no event at the pixel, it indicates that brightness of the pixel is not changed, and the event polarity includes increased brightness and decreased brightness.

Abstract: This application provides a chip, a dynamic vision sensor, and a method for outputting pixel information. The chip includes a pixel array and a processing circuit, and each pixel row in the pixel array includes at least two pixel groups. In the processing circuit, based on a change of a light intensity value of at least one pixel in a first pixel group, a first signal including row location information of the first pixel group in the pixel array is generated, and a second signal including column location information of a pixel row in which the first pixel group is located is generated. The first signal and the second signal are sent to a pixel output circuit, to enable the pixel output circuit to accurately output location information of the first pixel group whose light intensity value changes in the pixel array.

Abstract: This application provides a pixel of an image sensor and an image sensor. The pixel may include an optical-to-electrical conversion circuit and a composite measurement circuit. The optical-to-electrical conversion circuit may be configured to generate a first current based on incident light that is incident on the pixel. The composite measurement circuit is coupled to the optical-to-electrical conversion circuit. The pixel may have a first operating mode and a second operating mode. The composite measurement circuit may generate a first pulse signal based on the first current when the pixel operates in the first operating mode, where the first pulse signal represents the intensity information of the incident light. When the pixel operates in the second operating mode, the pixel receives a plurality of sampling signals, and generates a plurality of second pulse signals based on the first current and the plurality of sampling signals.

Abstract: In a model training method, a computing device generates an initial virtual human speaking video based on an audio data set and a person speaking video. The computing device trains a lip synchronization parameter generation model based on the audio data set and a lip synchronization training parameter used as a model training label. The computing device then uses the lip synchronization parameter generation model to generate a lip synchronization parameter based on input audio, and the lip synchronization parameter is used to drive a virtual human to speak. The computing device expands a data amount of the lip synchronization training parameter by using the initial virtual human speaking video.

Abstract: A rendering model training method and apparatus are provided. A related video rendering method and apparatus, a device, and a storage medium are also provided. The rendering model training method includes: obtaining a first video including a face of a target object; mapping a facial action of the target object in the first video based on a three-dimensional face model, to obtain a second video including a three-dimensional face; and training an initial rendering model by using the second video as an input of the initial rendering model and using the first video as supervision of an output of the initial rendering model, to obtain a target rendering model. The second video generated based on the three-dimensional face model is used as a sample for training the rendering model.

Abstract: The present disclosure relates to optical flow estimation methods and apparatuses. One example method includes obtaining a first image frame and a second image frame, where the first image frame and the second image frame are any two adjacent image frames in an image sequence, and the image sequence is obtained by photographing a target scene. A first event frame is obtained, where the first event frame is used to describe a luminance change of the target scene within a time period from the first image frame to the second image frame. A target optical flow is determined based on the first image frame, the second image frame, and the first event frame, where the target optical flow is an optical flow from the first image frame to a target moment, and the target moment is any moment between the first image frame and the second image frame.

Abstract: A method includes: detecting an abnormal touch event on a display screen of a wearable device; and enabling gesture recognition in response to the abnormal touch event when the wearable device is worn on an arm of a user such that the wearable device can be effectively controlled when there is a touch exception on the wearable device, so that interaction with the wearable device can be conveniently performed.

Abstract: A data processing method includes receiving an event data stream, where the event data stream includes at least a first event data item and a second event data item, the first event data item includes a first timestamp for obtaining the first event data item, the second event data item includes a second timestamp for obtaining the second event data item, and the second event data item is obtained most recently before obtaining the first event data item; and obtaining a compressed event data stream corresponding to the event data stream, where the compressed event data stream includes at least a first compressed event data item corresponding to the first event data item, and the first compressed event data item includes first time information.