Key point prediction method and device, electronic device and storage medium
A prediction method and key point technology, applied in the field of image recognition, can solve the problems of reduced prediction accuracy, inability to integrate time series features, and low prediction accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0122] Example 1
[0123] figure 1 A flowchart of a method for predicting key points provided in Embodiment 1 of the present invention. The method may be executed by a device for predicting key points provided in real time by the present invention, and the device may be implemented in software and / or hardware. For example, key points The predicting means may be implemented by a processor in which it is arranged to execute corresponding instructions stored in the memory. see figure 1 , the method includes:
[0124] S110. Acquire a target area in the video frame that includes the person to be identified.
[0125] The video to be processed includes at least one video frame, and each video frame includes a person to be identified. The target area refers to an area in the video frame that contains the person to be recognized. Optionally, the target area may be an original video frame, or a sub-image generated by performing image preprocessing on the original video frame. Exemp...
Example Embodiment
[0174] Embodiment 2
[0175] Figure 5 This is a flowchart of a method for predicting key points provided by the second embodiment of the present invention. On the basis of the above embodiment, a method for predicting key points by any neural network unit in the neural network is provided. Accordingly, the method specifically includes:
[0176] S210: Acquire a target area in the video frame that includes the person to be identified.
[0177] S220. Input the target region into any neural network unit in the neural network.
[0178] S230. Extract the contour feature map of the target area based on the first sub-network of any neural network unit.
[0179] S240. Combine the contour feature map, the key point heat map of the previous target area, and the standard center heat map in sequence to generate a combined feature map, and input the combined feature map into the long short-term memory sub-network of any neural network unit.
[0180] S250 , generating a memory feature m...
Example Embodiment
[0191] Embodiment 3
[0192] Figure 9 This is a flowchart of a method for predicting key points provided by the third embodiment of the present invention. On the basis of the above embodiment, a method for predicting key points by multiple neural network units in a neural network is provided. Accordingly, the method specifically includes:
[0193] S310: Acquire a target area in the video frame that includes the person to be identified.
[0194] S320: Group the target areas, and input the target areas in each group into the corresponding neural network units in the neural network, wherein the number of target areas in each group is the same as the number of neural network units in the neural network.
[0195] S330. Extract the contour feature map of the target area based on the first sub-network of the corresponding neural network unit.
[0196] S340 . Combine the contour feature map, the key point heat map of the previous target area, and the standard center heat map in se...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap