公开(公告)号：US11019275B2

公开(公告)日：2021-05-25

申请号：US16504978

申请日：2019-07-08

Applicant: Samsung Electronics Co., Ltd.

Inventor： Saikat Kumar Das ,  Pradeep Kumar Sindhagatta Krishnappa ,  Pranav Prakash Deshpande ,  Karthik Srinivasan

IPC: H04N5/235 ,  H04N5/14 ,  H04N9/31 ,  H04N9/04 ,  G06T3/40

Abstract: A method and apparatus for capturing images with a high degree of movements in a captured scene are provided. The method includes receiving, by an image sensor, a sequence of images including a plurality of RGB image frames and color event data, generating a plurality of pseudo color filter array (CFA) frames based on the color event data, and generating dynamic images by combining the plurality of RGB image frames and the plurality of pseudo CFA frames.

公开(公告)号：US11057641B1

公开(公告)日：2021-07-06

申请号：US17019609

申请日：2020-09-14

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Manjit Hota ,  Karthik Srinivasan ,  Puneet Pandey ,  Ayush Goel

IPC: H04N19/56 ,  H04N19/172 ,  H04N19/137 ,  H04N19/57

Abstract: Provided is a method of motion estimation for processing a video stream comprising a plurality of frames, the method including segmenting at least one frame, from among the plurality of frames, into a plurality of blocks, determining an event density factor for each block included in a frame, wherein the event density factor of the block corresponds to a number of events accumulated in the block across frames in a predetermined time duration, comparing the determined event density factor with a threshold value, estimating a motion vector of the block based on the comparison, and processing the block in the video stream based on the estimated motion vector of the block.

公开(公告)号：US11539896B2

公开(公告)日：2022-12-27

申请号：US16664750

申请日：2019-10-25

Applicant: Samsung Electronics Co., Ltd.

Inventor： Pradeep Kumar Sindhagatta Krishnappa ,  Pranav Prakash Deshpande ,  Karthik Srinivasan ,  Saikat Kumar Das ,  Alok Shankarlal Shukla

IPC: H04N5/235 ,  H04N9/04 ,  G06T7/13 ,  G06T7/20

Abstract: A method and an apparatus for capturing a high quality dynamic image by setting a different row-wise exposure value when capturing a scene are provided. The dynamic image capturing method includes: generating an image by pre-capturing a scene via an event sensor; generating event data from the image; determining a row-wise exposure value of the image based on the event data; and determining a row-wise readout priority order of the image based on the row-wise exposure value of the image.

公开(公告)号：US11050945B2

公开(公告)日：2021-06-29

申请号：US16504978

申请日：2019-07-08

Applicant: Samsung Electronics Co., Ltd.

Inventor： Saikat Kumar Das ,  Pradeep Kumar Sindhagatta Krishnappa ,  Pranav Prakash Deshpande ,  Karthik Srinivasan

IPC: H04N5/235 ,  H04N5/14 ,  H04N9/31 ,  H04N9/04 ,  G06T3/40

Abstract: A method and apparatus for capturing images with a high degree of movements in a captured scene are provided. The method includes receiving, by an image sensor, a sequence of images including a plurality of RGB image frames and color event data, generating a plurality of pseudo color filter array (CFA) frames based on the color event data, and generating dynamic images by combining the plurality of RGB image frames and the plurality of pseudo CFA frames.

公开(公告)号：US11263751B2

公开(公告)日：2022-03-01

申请号：US16666225

申请日：2019-10-28

Applicant: Samsung Electronics Co., Ltd.

Inventor： Sujoy Saha ,  Karthik Srinivasan ,  Sourabh Singh Yadav ,  Suhas Shantaraja Palasamudram ,  Venkat Ramana Peddigari ,  Pradeep Kumar Sk ,  Pranav P Deshpande ,  Akankshya Kar

IPC: G06T7/11 ,  G06K9/00 ,  G06K9/62 ,  G06T7/246 ,  G06V20/56

Abstract: A method for real-time semantic image segmentation using a monocular event-based sensor includes capturing a scene using a red, green, blue (RGB) sensor to obtain a plurality of RGB frames and an event sensor to obtain event data corresponding to each of the plurality of RGB frames, performing object labeling for objects in a first RGB frame among the plurality of RGB frames by identifying one or more object classes, obtaining an event velocity of the scene by fusing the event data corresponding to the first RGB frame and at least one subsequent RGB frame among the plurality of RGB frames, determining whether the event velocity is greater than a predefined threshold, and performing object labeling for objects in the at least one subsequent RGB frame based on the determination.

公开(公告)号：US11244464B2

公开(公告)日：2022-02-08

申请号：US16296932

申请日：2019-03-08

Applicant: Samsung Electronics Co., Ltd.

Inventor： Karthik Srinivasan ,  Pradeep Kumar Sindhagatta Krishnappa ,  Pranav Prakash Deshpande ,  Rituparna Sarkar

IPC: G06T7/50 ,  G06T7/20 ,  G06T3/40 ,  G06T5/50

Abstract: A method and apparatus are provided for performing depth estimation of an object in an image of a scene. The method includes capturing the image of a scene; obtaining, by a sensor, from the image, pixel intensity data and event data; generating an event depth map using the event data, wherein the event data includes event map data of the image and event velocity data of the image; and generating a depth map for the object in the image using the event depth map and the pixel intensity data.