US 12,112,495 B2
Depth data filtering method and apparatus, electronic device, and readable storage medium
Jian Kang, Dongguan (CN)
Assigned to GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP., LTD., Dongguan (CN)
Filed by GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP., LTD., Dongguan (CN)
Filed on Dec. 22, 2021, as Appl. No. 17/559,103.
Application 17/559,103 is a continuation of application No. PCT/CN2020/097464, filed on Jun. 22, 2020.
Claims priority of application No. 201910626650.X (CN), filed on Jul. 11, 2019.
Prior Publication US 2022/0114744 A1, Apr. 14, 2022
Int. Cl. G06K 9/00 (2022.01); G01B 11/22 (2006.01); G01S 7/4865 (2020.01); G01S 17/894 (2020.01); G06T 5/70 (2024.01); G06T 7/55 (2017.01); G06V 10/22 (2022.01); G06V 10/30 (2022.01); G06V 10/34 (2022.01); G06V 10/72 (2022.01); G06V 10/74 (2022.01)
CPC G06T 7/55 (2017.01) [G01B 11/22 (2013.01); G01S 7/4865 (2013.01); G01S 17/894 (2020.01); G06T 5/70 (2024.01); G06V 10/225 (2022.01); G06V 10/30 (2022.01); G06V 10/34 (2022.01); G06V 10/72 (2022.01); G06V 10/761 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/10028 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A depth data filtering method, comprising:
obtaining, for each pixel, a depth difference value between two consecutive frames of depth maps;
marking an area formed by pixels as a first environment change area, the depth difference value of the pixels being smaller than a predetermined absolute depth deviation;
marking an area formed by pixels as a second environment change area, the depth difference value of the pixels being greater than or equal to the predetermined absolute depth deviation;
generating a first similarity weight corresponding to each pixel in the first environment change area based on the depth difference value, an enlarged predetermined original smoothing coefficient, and a pixel depth deviation value of a current frame;
generating a second similarity weight corresponding to each pixel in the second environment change area based on the depth difference value, a reduced predetermined original smoothing coefficient, and the pixel depth deviation value of the current frame; and
filtering the first environment change area based on the first similarity weight and filtering the second environment change area based on the second similarity weight.