-
公开(公告)号:US20240095874A1
公开(公告)日:2024-03-21
申请号:US18513966
申请日:2023-11-20
申请人: Stats LLC
发明人: Long Sha , Sujoy Ganguly , Patrick Joseph Lucey
IPC分类号: G06T3/00 , G06F18/214 , G06N3/08 , G06T7/11 , G06T7/80 , G06V10/24 , G06V10/82 , G06V20/40 , G06V30/19 , G06V30/262 , H04N21/854
CPC分类号: G06T3/00 , G06F18/214 , G06N3/08 , G06T3/0093 , G06T7/11 , G06T7/80 , G06V10/24 , G06V10/82 , G06V20/49 , G06V30/19173 , G06V30/274 , H04N21/854 , G06T2207/10016 , G06T2207/20081 , G06T2207/20084 , G06T2207/30244
摘要: A system and method of calibrating a broadcast video feed are disclosed herein. A computing system retrieves a plurality of broadcast video feeds that include a plurality of video frames. The computing system generates a trained neural network, by generating a plurality of training data sets based on the broadcast video feed and learning, by the neural network, to generate a homography matrix for each frame of the plurality of frames. The computing system receives a target broadcast video feed for a target sporting event. The computing system partitions the target broadcast video feed into a plurality of target frames. The computing system generates for each target frame in the plurality of target frames, via the neural network, a target homography matrix. The computing system calibrates the target broadcast video feed by warping each target frame by a respective target homography matrix.
-
公开(公告)号:US20240087193A1
公开(公告)日:2024-03-14
申请号:US18274923
申请日:2022-01-19
CPC分类号: G06T11/60 , G06T3/0093 , G06T3/4038 , G06T17/00 , G06V10/56 , H04N1/60
摘要: A method and an apparatus for editing at least two sequences of pictures are provided. Editing the sequences comprises modifying at least one picture of one of the at least two sequences, based on a user input, and updating at least one picture of at least one other sequence of the at least two sequences, with the modification.
-
公开(公告)号:US20240087179A1
公开(公告)日:2024-03-14
申请号:US18462703
申请日:2023-09-07
发明人: Renqiang Min , Kai Li , Hans Peter Graf , Haomiao Ni
CPC分类号: G06T11/00 , G06T3/0093 , G06V20/46
摘要: Methods and systems for training a model include training an encoder in an unsupervised fashion based on a backward latent flow between a reference frame and a driving frame taken from a same video. A diffusion model is trained that generates a video sequence responsive to an input image and a text condition, using the trained encoder to determine a latent flow sequence and occlusion map sequence of a labeled training video.
-
公开(公告)号:US11922683B2
公开(公告)日:2024-03-05
申请号:US17299720
申请日:2020-01-20
申请人: WheelRight Limited
IPC分类号: G06K9/00 , G06F18/213 , G06F18/214 , G06F18/24 , G06N3/08 , G06T3/00 , G06V10/82 , G06V10/94 , G06V20/62 , G06V30/146 , G06V30/18 , G06V30/10
CPC分类号: G06V10/82 , G06F18/213 , G06F18/214 , G06F18/24 , G06N3/08 , G06T3/0093 , G06V10/95 , G06V20/63 , G06V30/147 , G06V30/18057 , G06V30/18095 , G06V30/10
摘要: A computer implemented method for generating a region of interest on a digital image of a sidewall of a tyre, the sidewall having one or more embossed and/or engraved markings, is provided. The method comprises generating a histogram of oriented gradients feature map of the digital image, inputting the histogram of oriented gradients feature map into a trained convolutional neural network, wherein said trained convolutional neural network is configured to output a first probability based on the input histogram of oriented gradients feature map that a region of pixels of the digital image contains the embossed and/or engraved markings, and if the first probability is at or above a first predetermined threshold, accepting said region of pixels as said region of interest.
-
公开(公告)号:US11915384B2
公开(公告)日:2024-02-27
申请号:US17237508
申请日:2021-04-22
发明人: Jun Murakawa , Sagnik Chowdhury
CPC分类号: G06T3/40 , G06T3/0093
摘要: A technique iteratively fits a rasterization distortion function to a distortion property of a display such as a head-mounted display (HMD). The technique also features reconstructing a distortion function from a rendered image with unknown distortion property.
-
公开(公告)号:US20240062328A1
公开(公告)日:2024-02-22
申请号:US17820587
申请日:2022-08-18
IPC分类号: G06T3/00
CPC分类号: G06T3/0093
摘要: Aspects of the invention include systems and methods configured for hyper-personalized feature morphing of an avatar within a metaverse. A non-limiting example computer-implemented method includes predicting a next interaction for the avatar and comparing one or more current features of the avatar to one or more feature requirements for the predicted next interaction. The method further includes determining, based on the comparison, that one or more current features of the avatar do not match one or more feature requirements for the predicted next interaction. Responsive to the determination, features of the avatar are altered to satisfy the one or more feature requirements for the predicted next interaction.
-
公开(公告)号:US11908109B1
公开(公告)日:2024-02-20
申请号:US17089350
申请日:2020-11-04
申请人: madVR Holdings, LLC
发明人: Mathias Rauen
CPC分类号: G06T5/009 , G06T3/0093 , G06T3/40 , G06T5/002 , G06T5/50 , G06T9/00 , G06T2207/10016 , G06T2207/20024 , G06T2207/20192 , G06T2207/20208
摘要: Various improvements to video processing are described, including highlight recovery by dividing an image into different frequency ranges and compressing the lower range before recombining, tone mapping to avoid hue shifts, dynamic tone mapping based on detected scene changes in the video, dynamic tone mapping using the shape of a tone mapping curve that is changed based on the histogram of each video frame, HDR Luminance Channel Repair of artifacts, changing the shape of the tone mapping curve based on a histogram of each video frame, upscaling Chroma by using Luma channel information motion interpolation using Neural Networks, motion compensated noise reduction using Neural Networks to filter random noise and film grain, grain/noise agnostic upscaling using Neural Networks, and hiding video frames by strategically dropping or repeating them in unnoticeable places to prevent a visible video stutter based on scene characteristics to accommodate clock variations.
-
公开(公告)号:US11908100B2
公开(公告)日:2024-02-20
申请号:US17201660
申请日:2021-03-15
发明人: Shizhong Liu , Jincheng Huang , Weiliang Liu , Jianfeng Ren , Sahul Madanayakanahalli Phaniraj Venkatesh
CPC分类号: G06T3/0093 , G06T3/0012 , G06T5/50 , G06T7/521 , G06T7/571 , H04N5/268 , H04N23/90 , G06T2207/10016 , G06T2207/10028
摘要: Artefacts in a sequence of image frames may be reduced or eliminated through modification of an input image frame to match another image frame in the sequence, such as by geometrically warping to generate a corrected image frame with a field of view matched to another frame in sequence of frames with the image frame. The warping may be performed based on a model generated from data regarding the multi-sensor device. The disparity between image frames may be modeled based on image captures from the first and second image sensor for scenes at varying depths. The model may be used to predict disparity values for captured images, and those predicted disparity values used to reduce artefacts resulting from image sensor switching. The predicted disparity values may be used in image conditions resulting in erroneous actual disparity values.
-
公开(公告)号:US20240054615A1
公开(公告)日:2024-02-15
申请号:US17887157
申请日:2022-08-12
申请人: Meta Platforms, Inc.
发明人: Lucas Young , Zhuo Hui , Denis Demandolx , Rakesh Ranjan , Yuchen Fan
CPC分类号: G06T5/002 , G06T3/0093 , G06T5/50 , G06T7/38 , G06T7/0002 , G06T2207/10024 , G06T2207/30168 , G06T2207/20081 , G06T2207/20084
摘要: In one embodiment, a method includes accessing a sequence of raw images comprising at least a first raw image, a second raw image, and a third raw image sequentially, wherein the second raw image comprises image noise, warping the first and third raw images with respect to the second raw image based on an optical flow associated with the sequence of raw images, generating an input tensor based on the first warped raw image, the second raw image, and the third warped raw image, and generating a denoised raw image based on one or more machine-learning models for the second raw image based on the input tensor.
-
公开(公告)号:US11902492B2
公开(公告)日:2024-02-13
申请号:US17265119
申请日:2019-08-01
发明人: Seung-Ho Jung , Hocheol Shin , Chang-Hoi Kim , You-Rak Choi , Dongseok Ryu , Jonghui Han
IPC分类号: H04N13/128 , H04N13/239 , G06T3/00 , G06T3/60
CPC分类号: H04N13/128 , G06T3/0093 , G06T3/60 , H04N13/239
摘要: An image processing method includes: calculating a rotation matrix to eliminate vertical parallax by converting stereo image captured by stereo camera as if it is image captured by camera rotated about an origin of camera coordinate system, wherein the stereo image includes left image obtained by left camera and right image obtained by right camera; calculating a rotation matrix to adjust the horizontal parallax of the left image and a rotation matrix to adjust the horizontal parallax of the right image; calculating a geometric transformation matrix for the stereo image based on each camera characteristic matrix representing lens characteristics, the rotation matrix for eliminating vertical parallax, and the rotation matrices for adjusting horizontal parallax; and generating new stereo image in which vertical parallax is eliminated and horizontal parallax is adjusted by performing image warping for the stereo image captured by the stereo camera based on the calculated geometric transformation matrix.
-
-
-
-
-
-
-
-
-