US 11,715,180 B1
Emirror adaptable stitching
I-Hsuan Chen, New Taipei (TW); Hung-Ling Lu, New Taipei (TW); and Yen-Yu Chen, Taipei (TW)
Assigned to Ambarella International LP, Santa Clara, CA (US)
Filed by Ambarella International LP, Santa Clara, CA (US)
Filed on Feb. 9, 2022, as Appl. No. 17/667,640.
Application 17/667,640 is a continuation of application No. 16/580,222, filed on Sep. 24, 2019, granted, now 11,341,614.
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 5/00 (2006.01); B60R 1/12 (2006.01); G06T 7/50 (2017.01); B60R 1/00 (2022.01); G06T 7/73 (2017.01); G06T 3/40 (2006.01)
CPC G06T 5/002 (2013.01) [B60R 1/00 (2013.01); B60R 1/12 (2013.01); G06T 3/4038 (2013.01); G06T 7/50 (2017.01); G06T 7/73 (2017.01); B60R 2001/1215 (2013.01); B60R 2300/105 (2013.01); B60R 2300/303 (2013.01); B60R 2300/8066 (2013.01); G06T 2207/10012 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/30252 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An apparatus comprising:
an interface configured to receive pixel data from a left capture device, a right capture device and a center capture device; and
a processor configured to (a) process (i) said pixel data from said left capture device arranged as left video frames, (ii) said pixel data from said right capture device arranged as right video frames and (iii) said pixel data from said center capture device as center video frames, (b) perform computer vision operations to detect objects in said center video frames, (c) determine monocular depth information in response to said objects detected in said center video frames, (d) determine adjustments to blending lines in response to a comparison of said monocular depth information to a missing object range for stitching said center video frames to both (a) said left video frames and (b) said right video frames, (e) perform video stitching operations based on said blending lines on (i) said left video frames and said center video frames and (ii) said right video frames and said center video frames and (f) generate stitched video frames in response to said video stitching operations, wherein said blending lines are determined to prevent said objects from being in said missing object range.