Multi-sensor video camera, and a method and processing pipeline for the same

A processing pipeline for a multi-sensor video camera having a first and a second sensor with partly overlapping fields of view, comprising:

• a receiver configured to receive a first video frame and a second video frame being simultaneously captured by the first sensor and the second sensor, respectively,
• wherein each of the first and the second video frame has a non-overlapping portion where it does not overlap with the other of the first and the second video frame, and an overlapping portion where it overlaps with the other of the first and the second video frame;
• a frame generating component configured to generate a frame of a first video stream by joining together image data from the non-overlapping portions of the first and the second video frame with image data from the overlapping portion of the first video frame only, and to generate a frame of a second video stream, wherein the frame of the second video stream includes image data from the overlapping portion of at least the second video frame,
• a first processing component configured to process the frame of the first video stream,
• a second processing component configured to process the frame of the second video stream in parallel with the first processing component processing the frame of the first video stream, wherein the processing of the frame of the second video stream includes preparing an overlay based on the image data from the overlapping portion of at least the second video frame;
• and
• an overlay component configured to add the overlay to the processed frame of the first video stream at a portion thereof corresponding to the overlapping portion of the first video frame,
• wherein at least one of the second processing component and the overlay component is configured to blend image data from the overlapping portion of the first video frame with image data from the overlapping portion of the second video frame when preparing the overlay or adding the overlay.

A multi-sensor video camera, comprising:

• a first sensor arranged to capture video frames depicting a first field of view;
• a second sensor arranged to capture video frames simultaneously with the first sensor, the second sensor depicting a second field of view which partially overlaps with the first field of view;
• and
• a processing pipeline comprising:
• a receiver configured to receive a first video frame and a second video frame being simultaneously captured by the first sensor and the second sensor, respectively,
• wherein each of the first and the second video frame has a non-overlapping portion that does not overlap with the other of the first and the second video frame, and an overlapping portion that overlaps with the other of the first and the second video frame;
• a frame generating component configured to generate a frame of a first video stream by joining together image data from the non-overlapping portions of the first and the second video frame with image data from the overlapping portion of the first video frame only, and to generate a frame of a second video stream, wherein the frame of the second video stream includes image data from the overlapping portion of at least the second video frame,
• a first processing component configured to process the frame of the first video stream,
• a second processing component configured to process the frame of the second video stream in parallel with the first processing component processing the frame of the first video stream, wherein the processing of the frame of the second video stream includes preparing an overlay based on the image data from the overlapping portion of at least the second video frame;
• and
• an overlay component configured to add the overlay to the processed frame of the first video stream at a portion thereof corresponding to the overlapping portion of the first video frame,
• wherein at least one of the second processing component and the overlay component is configured to blend image data from the overlapping portion of the first video frame with image data from the overlapping portion of the second video frame when preparing the overlay or adding the overlay.

A computer program product comprising a computer-readable medium having computer-code instructions stored thereon for carrying out a method that, when executed by a device having processing capability, cause the device to:

• receive a first video frame and a second video frame being simultaneously captured by the first sensor and the second sensor, respectively,
• wherein each of the first and the second video frame has a non-overlapping portion that does not overlap with the other of the first and the second video frame, and an overlapping portion that overlaps with the other of the first and the second video frame,
• generate a frame of a first video stream by joining together image data from the non-overlapping portions of the first and the second video frame with image data from the overlapping portion of the first video frame only,
• generate a frame of a second video stream, wherein the frame of the second video stream includes image data from the overlapping portion of at least the second video frame,
• process the frame of the first video stream and the frame of the second video stream in parallel, wherein the processing of the frame of the second video stream includes preparing an overlay based on the image data from the overlapping portion of at least the second video frame, and
• add the overlay to the processed frame of the first video stream at a portion thereof corresponding to the overlapping portion of the first video frame,
• wherein image data from the overlapping portion of the first video frame is blended with image data from the overlapping portion of the second video frame when preparing the overlay or adding the overlay.