Video processing method and associated system on chip
The present invention provides a SoC including a recognition circuit and a processing circuit. The recognition circuit is configured to obtain image data from an image capturing device, and perform a recognition operation on the image data to generate a recognition result. The processing circuit is coupled to the recognition circuit, and is configured to determine a ROI in the image data according to the recognition result, perform image enhancement operation on the ROI to generate an enhanced region, and combine the enhanced region with the image data to generate processed image data.
1. A system on chip (SoC), comprising:
a recognition circuit, configured to obtain image data from an image capturing device, and perform a recognition operation on the image data to generate a recognition result; and
a processing circuit, coupled to the recognition circuit, configured to determine a region of interest (ROI) in the image data according to the recognition result, perform image enhancement operation on the ROI to generate an enhanced region, and combine the enhanced region with the image data to generate processed image data;
wherein the recognition circuit is a person recognition circuit, the person recognition circuit performs a person recognition operation on the image data to generate the recognition result, and the SoC further comprises:
a sound detection circuit, configured to receive a plurality of sound signals from a plurality of microphones, and detect a position/direction of a main sound to generate a sound detection result;
wherein the processing circuit determines a region where a speaker is located in the image data according to the recognition result and the sound detection result, as the ROI.
2. The SoC of claim 1 , wherein the processing circuit covers the enhanced region to a specific area of the image data to generate the processed image data.
3. The SoC of claim 2 , wherein the specific area does not overlap the ROI.
4. The SoC of claim 2 , wherein the processing circuit performs an enlargement operation and a resolution enhancement operation on the ROI to generate the enhanced region.
5. The SoC of claim 1 , wherein the recognition result comprises a plurality of regions, each region comprises a person, and the processing circuit refers to the recognition result and the sound detection result to select one of the regions to serve as the ROI.
6. The SoC of claim 1 , wherein the SoC is used in an electronic device, and the processed image data is transmitted from the electronic device to another electronic device via network.
7. A video processing method, comprising:
obtaining image data from an image capturing device, and performing a recognition operation on the image data to generate a recognition result;
determining a region of interest (ROI) in the image data according to the recognition result;
performing image enhancement operation on the ROI to generate an enhanced region; and
combining the enhanced region with the image data to generate processed image data;
wherein the step of performing the recognition operation on the image data to generate the recognition result is to perform a person recognition operation on the image data to generate the recognition result, and the video processing method further comprises:
receiving a plurality of sound signals from a plurality of microphones, and detecting a position/direction of a main sound to generate a sound detection result; and
the step of determining the RIO in the image data according to the recognition result comprises:
determines a region where a speaker is located in the image data according to the recognition result and the sound detection result, as the ROI.
8. The video processing method of claim 7 , wherein the step of combining the enhanced region with the image data to generate the processed image data comprises:
covering the enhanced region to a specific region of the image data to generate the processed image data.