Abstract: Examples of video encoding methods and apparatus and video decoding methods and apparatus are described. An example method of processing video data includes determining, for a conversion between a video block of a video and a bitstream of the video, a gradient of a prediction vector at a sub-block level for the video block according to a rule, wherein the rule specifies to use a same gradient value is assigned for all samples within a sub-block of the video block; and performing the conversion based on the determining.

Abstract: Examples of video encoding methods and apparatus and video decoding methods and apparatus are described. An example method of processing video data includes determining, for a conversion between a video block of a video and a bitstream of the video, a gradient of a prediction vector at a sub-block level for the video block according to a rule, wherein the rule specifies to use a same gradient value is assigned for all samples within a sub-block of the video block; and performing the conversion based on the determining.

Abstract: An earbud, including a housing, a speaker assembly, and a partition, is provided. The housing includes a sound outlet hole and a sound tuning hole, wherein the sound outlet hole is in acoustic communication with an ear canal of a human ear but does not contact the ear canal. The speaker assembly is disposed in the housing to separate a space in the housing into a front cavity and a rear cavity. The partition is disposed in the front cavity of the housing to divide the front cavity into a first area close to the speaker assembly and a second area away from the speaker assembly. The partition is between the first area and the second area, and the partition has a thickness and includes multiple first through holes, wherein the first through holes pass through the partition to connect the first area with the second area.

Abstract: An earbud, including a housing, a speaker assembly, and a partition, is provided. The housing includes a sound outlet hole and a sound tuning hole, wherein the sound outlet hole is in acoustic communication with an ear canal of a human ear but does not contact the ear canal. The speaker assembly is disposed in the housing to separate a space in the housing into a front cavity and a rear cavity. The partition is disposed in the front cavity of the housing to divide the front cavity into a first area close to the speaker assembly and a second area away from the speaker assembly. The partition is between the first area and the second area, and the partition has a thickness and includes multiple first through holes, wherein the first through holes pass through the partition to connect the first area with the second area.

Abstract: An image processing method, apparatus, and device, and a storage medium relate to the field of artificial intelligence. The method may include: obtaining an image corresponding to a target battery module, the target battery module including N solder joints, and the N solder joints being respectively mapped to N solder joint fields in the image; dividing the image according to the N solder joint fields to obtain N image regions in one-to-one correspondence with the N solder joint fields; calculating image difference information between each pair of adjacent image regions among the N image regions to obtain an image difference information set; and performing fault recognition on the target battery module based on the image difference information set. The accuracy for recognizing a preset fault in a battery module can be improved by the method.

Abstract: This application relates to the field of artificial intelligence, and specifically, to the field of computer vision. The method includes: performing feature extraction on an image to obtain a basic feature map of the image; determining a proposal of a region possibly including a pedestrian in the image; processing the basic feature map of the image to obtain an object visibility map in which a response to a pedestrian visible part is greater than a response to a pedestrian blocked part and a background part; performing weighted summation processing on the object visibility map and the basic feature map to obtain an enhanced feature map of the image; and determining, based on the proposal of the image and the enhanced feature map of the image, a bounding box including a pedestrian in the image and a confidence level of the bounding box including the pedestrian in the image.

Abstract: Provided are an audio calibration method and device. The method includes: transmitting an audio calibration signal to a first loudspeaker and a second loudspeaker such that the first loudspeaker and the second loudspeaker generate sound signals; determining first loudspeaker sound information through a sound signal generated by the first loudspeaker and acquired by a first microphone and a sound signal generated by the first loudspeaker and acquired by the second microphone; determining second loudspeaker sound information through a sound signal generated by the second loudspeaker and acquired by the first microphone and a sound signal generated by the second loudspeaker and acquired by the second microphone; and determining a calibration parameter of the first loudspeaker or a calibration parameter of the second loudspeaker according to the first loudspeaker sound information and the second loudspeaker sound information.

Abstract: Provided are a wireless earphone and a method for controlling a wireless earphone. The wireless earphone includes a charging box and an earpiece placed in the charging box. A charging box open/closed-state sensor, an earpiece attitude sensor and a magnetic-field control circuit are disposed in the charging box. The charging box open/close-state sensor is configured to detect an open state of the charging box or a closed state of the charging box. The earpiece attitude sensor is configured to detect an orientation attitude of the earpiece in the charging box. The magnetic-field control circuit controls, according to results detected by the charging box open/closed-state sensor and the earpiece attitude sensor, whether the earpiece is suspended from the charging box.

Abstract: Provided are an audio calibration method and device. The method includes: transmitting an audio calibration signal to a first loudspeaker and a second loudspeaker such that the first loudspeaker and the second loudspeaker generate sound signals; determining first loudspeaker sound information through a sound signal generated by the first loudspeaker and acquired by a first microphone and a sound signal generated by the first loudspeaker and acquired by the second microphone; determining second loudspeaker sound information through a sound signal generated by the second loudspeaker and acquired by the first microphone and a sound signal generated by the second loudspeaker and acquired by the second microphone; and determining a calibration parameter of the first loudspeaker or a calibration parameter of the second loudspeaker according to the first loudspeaker sound information and the second loudspeaker sound information.

Abstract: Provided are a wireless earphone and a method for controlling a wireless earphone. The wireless earphone includes a charging box and an earpiece placed in the charging box. A charging box open/closed-state sensor, an earpiece attitude sensor and a magnetic-field control circuit are disposed in the charging box. The charging box open/close-state sensor is configured to detect an open state of the charging box or a closed state of the charging box. The earpiece attitude sensor is configured to detect an orientation attitude of the earpiece in the charging box. The magnetic-field control circuit controls, according to results detected by the charging box open/closed-state sensor and the earpiece attitude sensor, whether the earpiece is suspended from the charging box.