The architecture shown can perform global search, local search and local sub pixel search in a parallel or in a pipelined mode. All operations are in a streaming mode without the requirement of external intermediate data storage.

In an object tracking device, an extraction means extracts target candidates from time series images. A search range update means updates a search range based on frame information of a target in a previous image in a time series and a movement pattern of the target. A tracking means searches for and tracks the target using a confidence level indicating similarity with a target model among the target candidates. A model update means updates the target model using the target candidates extracted in the search range.

A vehicle external environment recognition apparatus to be applied to a vehicle includes one or more processors and one or more memories configured to be coupled to the one or more processors. The one or more processors are configured to: calculate three-dimensional positions of respective blocks in a captured image; group the blocks to put any two or more of the blocks that have the three-dimensional positions differing from each other within a predetermined range in a group and thereby determine three-dimensional objects; identify each of a preceding vehicle of the vehicle and a sidewall on the basis of the determined three-dimensional objects; and track the preceding vehicle. The one or more processors are configured to determine, upon tracking the preceding vehicle, whether the preceding vehicle to track is to be hidden by the sidewall on the basis of a border line between a blind region and a viewable region.

An image processing apparatus includes an image acquisition unit configured to extract an image from a video image, a history management unit configured to generate movement history of an object, a setting unit configured to set a search range in the image based on the movement history, and a tracking unit configured to detect a target object in the search range and associate a currently detected target object with a previously detected target object, to perform tracking on the target object.

The present application provides a volumetric measuring method and apparatus based on a TOF depth camera, which is achieved by: acquiring depth images; denoising the depth images; selecting a first feature area of the first depth image and a second feature area of the second depth image that have been denoised respectively, traversing each pixel point in the first feature area at multiple traversal speeds, and matching a first depth value of a current pixel point with a second depth value of a target pixel point corresponding to the current pixel point in the second feature area to acquire multiple total similarity values; determining a traversal speed corresponding to a maximum total similarity value as an instantaneous speed at a current moment; calculating a package volume according to the instantaneous speed. The present application effectively solves the problem that an accurate package volume cannot be obtained when a speed of the conveyor belt is not fixed, and high in feasibility.

The present application provides a volumetric measuring method and apparatus based on a TOF depth camera, which is achieved by: acquiring depth images; denoising the depth images; selecting a first feature area of the first depth image and a second feature area of the second depth image that have been denoised respectively, traversing each pixel point in the first feature area at multiple traversal speeds, and matching a first depth value of a current pixel point with a second depth value of a target pixel point corresponding to the current pixel point in the second feature area to acquire multiple total similarity values; determining a traversal speed corresponding to a maximum total similarity value as an instantaneous speed at a current moment; calculating a package volume according to the instantaneous speed. The present application effectively solves the problem that an accurate package volume cannot be obtained when a speed of the conveyor belt is not fixed, and high in feasibility.

A vehicle external environment recognition apparatus to be applied to a vehicle includes one or more processors and one or more memories configured to be coupled to the one or more processors. The one or more processors are configured to: calculate three-dimensional positions of respective blocks in a captured image; group the blocks to put any two or more of the blocks that have the three-dimensional positions differing from each other within a predetermined range in a group and thereby determine three-dimensional objects; identify each of a preceding vehicle of the vehicle and a sidewall on the basis of the determined three-dimensional objects; and track the preceding vehicle. The one or more processors are configured to determine, upon tracking the preceding vehicle, whether the preceding vehicle to track is to be hidden by the sidewall on the basis of a border line between a blind region and a viewable region.

Methods and system, including decoders and encoders, for interprediction. In one aspect, a method includes selecting reference samples based on motion information of a current picture block of a current picture, deriving first interpolated samples by performing a first interpolation on the selected reference samples, deriving an integer distance delta motion vector for a target sub-prediction unit (PU) by performing integer-distance MVR, deriving M×M pixel matrix flow vectors by performing BPOF, for each M×M pixel matrix in the target sub-PU, based on the first interpolated samples and the integer distance delta motion vector, deriving second interpolated samples by performing a second interpolation on the reference samples, computing at least one correction parameter for the target sub-PU based on the M×M pixel matrix flow vectors, the first interpolated samples and the second interpolated samples, and performing bi-prediction based on the second interpolated samples and the at least one correction parameter.

An image processing method includes: downsizing a current frame and a reference frame; dividing the down-sized current frame and the down-sized reference frame into multiple first current blocks and multiple first reference blocks, respectively; performing a first motion estimation to the first current blocks and the first reference blocks to generate multiple first motion vectors; dividing the current picture and the reference picture into multiple second current blocks and multiple second reference blocks, respectively; performing a second motion estimation to the second current blocks and the second reference blocks to generate multiple second motion vectors; and generating a compensated frame between the current frame and the reference frame according to the second motion vectors. The second motion estimation includes: performing a 3D recursive search for each second current block; and adjusting multiple estimation parameters in the 3D recursive search according to the first motion vector.

A method includes receiving a reference image and a non-reference image; dividing the reference image into a plurality of tiles; determining, using an electronic device, a motion vector map using coarse-to-fine based motion vector estimation; and generating an output frame using the motion vector map with the reference image and the non-reference image.