Depth detection method, device, electronic device and computer readable storage medium
Through image segmentation and key point detection, combined with human body mask image, the target human body depth detection in the image without a three-dimensional depth camera is achieved, solving the problem of depth detection when the existing technology lacks special hardware equipment, and is suitable for AR and virtual photography applications improve the accuracy and flexibility of detection.
Patent Information
- Application Number
- CN202011344694.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-11-24
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2040-11-24
AI Technical Summary
In the absence of special hardware devices such as three-dimensional depth cameras, it is difficult to effectively solve the problem of how to achieve single-person depth detection in images, especially in applications such as augmented reality (AR) interaction and virtual photography.
By acquiring the images collected by the image acquisition device, segmenting a single human body image, detecting two-dimensional and three-dimensional key point information, combining the human body mask image, determining the depth detection results of the target human body, and using optimization processing and filtering technology to ensure the accuracy of depth detection. sex.
Without relying on a three-dimensional depth camera, the depth detection of the target human body in the image is realized, which is suitable for scenes such as AR interaction and virtual photography, improving the accuracy of detection and application flexibility.
Smart Images

Figure CN112419388B_ABST
