Matching cost processing method and circuit and cost aggregation processing method
A technology of matching costs and processing methods, applied in the field of artificial intelligence, can solve problems such as complex system configuration, lower performance requirements, and high cost, and achieve the effects of efficient data reuse, improved flexibility, and reduced data storage
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
example 2
[0054] Example 2, the matching cost algorithm of gray-level cross-correlation, can input the vector of the reference pixel and the vector of the corresponding target pixel to the initial cost calculation circuit at one time, and the core of the similarity calculation circuit can be the multiply-add circuit.
example 3
[0055] Example 3, normalized grayscale cross-correlation algorithm, one vector can be subjected to dot product multiplication and accumulation operation, and each vector element of the other channel is first squared and then accumulated. After the accumulation of each window vector is completed, multiplication and square root operation are performed , the core of the similarity calculation circuit may include an accumulation circuit, a multiplier, a square root operation circuit, and the like.
[0056] The above-mentioned similarity calculation circuit is a schematic illustration, and the specific circuit form can be selected according to the similarity calculation algorithm, which is not limited here.
[0057] In implementation, an initial cost parallel computing module may be set on the deployment application side (that is, the application side where the matching cost processing is deployed), wherein the initial cost parallel computing module includes a first register group, ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


