公开(公告)号：US11665358B2

公开(公告)日：2023-05-30

申请号：US16860758

申请日：2020-04-28

Applicant: Adobe Inc.

Inventor： Viswanathan Swaminathan ,  Stefano Petrangeli ,  Gwendal Simon

IPC: H04N19/00 ,  H04N19/192 ,  H04N19/186 ,  H04N19/176

CPC classification number: H04N19/192 ,  H04N19/176 ,  H04N19/186

Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media to enhance texture image delivery and processing at a client device. For example, the disclosed systems can utilize a server-side compression combination that includes, in sequential order, a first compression pass, a decompression pass, and a second compression pass. By applying this compression combination to a texture image at the server-side, the disclosed systems can leverage both GPU-friendly and network-friendly image formats. For example, at a client device, the disclosed system can instruct the client device to execute a combination of decompression-compression passes on a GPU-network-friendly image delivered over a network connection to the client device. In so doing, client device can generate a tri-pass-compressed-texture from a decompressed image comprising texels with color palettes based on previously reduced color palettes from the first compression pass at the server-side, which reduces computational overhead and increases performance speed.

公开(公告)号：US20220343155A1

公开(公告)日：2022-10-27

申请号：US17337998

申请日：2021-06-03

Applicant: Adobe Inc.

Inventor： Saayan Mitra ,  Gang Wu ,  Georgios Theocharous ,  Richard Whitehead ,  Viswanathan Swaminathan ,  Zahraa Parekh ,  Ben Tepfer

IPC: G06N3/08 ,  G06K9/62 ,  G06F9/48

Abstract: This disclosure describes methods, non-transitory computer readable storage media, and systems that intelligently generate and modify schedules of task sequences utilizing a graph neural network and/or reinforcement learning model. For example, the disclosed system utilizes a graph neural network to generate performance efficiency scores indicating predicted performances of the sets of tasks. Additionally, the disclosed systems utilizes the performance efficiency scores to rank sets of tasks and then determine a schedule including an ordered sequence of tasks. Furthermore, disclosed system generates modified schedules in response to detecting a modification to the schedule. For example, the disclosed system utilizes a reinforcement learning model to provide recommendations of new tasks or task sequences deviating from the schedule in the event of an interruption. The disclosed system also utilizes the reinforcement learning model to learn from user choices to inform future scheduling of tasks.

公开(公告)号：US11457263B2

公开(公告)日：2022-09-27

申请号：US16784100

申请日：2020-02-06

Applicant: Adobe Inc.

Inventor： Viswanathan Swaminathan ,  Mohammad Hosseini

IPC: H04N21/2662 ,  H04L47/80 ,  H04N21/81 ,  H04N21/462 ,  H04L65/80 ,  H04N21/845 ,  H04L65/70 ,  H04L65/612 ,  H04L65/75

Abstract: The present disclosure includes methods and systems for streaming high-performance virtual reality video using adaptive rate allocation. In particular, an adaptive rate allocation system partitions a panorama video into segments or tiles and assigns priorities to each tile or segment based on input (e.g., a viewport of field-of-view) from a user client device. Further, the adaptive rate allocation system streams each tile or segment to the user client device according to the adaptive rate allocation, which maximizes bandwidth efficiency and video quality. In this manner, the adaptive rate allocation system delivers higher quality content to regions in the panorama video where a user is currently looking/most likely to look.

公开(公告)号：US20220245446A1

公开(公告)日：2022-08-04

申请号：US17164111

申请日：2021-02-01

Applicant: ADOBE INC.

Inventor： Saayan Mitra ,  Xiang Chen ,  Akangsha Sunil Bedmutha ,  Viswanathan Swaminathan ,  Omar Rahman ,  Camille Girabawe

IPC: G06N3/08 ,  G06N3/04

Abstract: An improved electronic communication system schedules transmission of electronic communications based on a predicted open time and click time. The open and click times are predicted from a machine learning model that is trained to optimize for both tasks. Additionally, when training the machine learning model, the loss used for adjusting the system to achieve a desired accuracy may be a biased loss determined from a function that penalizes overpredicting the open time. As such, the loss value may be determined by different set of rules depending on whether the predicted time is greater than the actual time or not.

公开(公告)号：US20220222866A1

公开(公告)日：2022-07-14

申请号：US17148928

申请日：2021-01-14

Applicant: Adobe Inc.

Inventor： Stefano Petrangeli ,  Viswanathan Swaminathan ,  Haoliang Wang

IPC: G06T9/40 ,  G06T3/40 ,  H04N19/96 ,  H04N19/13 ,  H04N19/186 ,  H04N19/169 ,  G06N7/00

Abstract: In implementations of systems for digital image compression using context-based pixel predictor selection, a computing device implements a compression system to receive digital image data describing pixels of a digital image. The compression system groups first differences between values of the pixels and first prediction values of the pixels into context groups. A pixel predictor is determined for each of the context groups based on a compression criterion. The compression system generates second prediction values of the pixels using the determined pixel predictor for pixels corresponding to the first differences included in each of the context groups. Second differences between the values of the pixels and the second prediction values of the pixels are grouped into different context groups. The compression system compresses the digital image using entropy coding based on the different context groups.

公开(公告)号：US20220156886A1

公开(公告)日：2022-05-19

申请号：US17097600

申请日：2020-11-13

Applicant: ADOBE INC.

Inventor： Stefano Petrangeli ,  Viswanathan Swaminathan ,  Haoliang Wang ,  YoungJoong Kwon

IPC: G06T3/60 ,  G06N3/08 ,  G06N3/04

Abstract: Methods, system, and computer storage media are provided for novel view synthesis. An input image depicting an object is received and utilized to generate, via a neural network, a target view image. In exemplary aspects, additional view images are also generated within the same pass of the neural network. A loss is determined based on the target view image and additional view images and is used to modify the neural network to reduce errors. In some aspects, a rotated view image is generated by warping a ground truth image from an initial angle to a rotated view angle that matches a view angle of an image synthesized via the neural network, such as a target view image. The rotated view image and the synthesized image matching the rotated view angle (e.g., a target view image) are utilized to compute a rotational loss.

公开(公告)号：US20220156503A1

公开(公告)日：2022-05-19

申请号：US16953049

申请日：2020-11-19

Applicant: Adobe Inc.

Inventor： Viswanathan Swaminathan ,  Stefano Petrangeli ,  Hongxiang Gu

IPC: G06K9/00 ,  G11B27/06 ,  G11B27/031 ,  G06K9/62 ,  G06N3/08

Abstract: A video summarization system generates a concatenated feature set by combining a feature set of a candidate video shot and a summarization feature set. Based on the concatenated feature set, the video summarization system calculates multiple action options of a reward function included in a trained reinforcement learning module. The video summarization system determines a reward outcome included in the multiple action options. The video summarization system modifies the summarization feature set to include the feature set of the candidate video shot by applying a particular modification indicated by the reward outcome. The video summarization system identifies video frames associated with the modified summarization feature set, and generates a summary video based on the identified video frames.

公开(公告)号：US20210272341A1

公开(公告)日：2021-09-02

申请号：US16804822

申请日：2020-02-28

Applicant: Adobe Inc.

Inventor： Viswanathan Swaminathan ,  Gang Wu ,  Akshay Malhotra

IPC: G06T11/60 ,  G06K9/62 ,  G06T7/11 ,  G06T7/70 ,  G06K9/00 ,  G06N3/08

Abstract: Generating images and videos depicting a human subject wearing textually defined attire is described. An image generation system receives a two-dimensional reference image depicting a person and a textual description describing target clothing in which the person is to be depicted as wearing. To maintain a personal identity of the person, the image generation system implements a generative model, trained using both discriminator loss and perceptual quality loss, which is configured to generate images from text. In some implementations, the image generation system is configured to train the generative model to output visually realistic images depicting the human subject in the target clothing. The image generation system is further configured to apply the trained generative model to process individual frames of a reference video depicting a person and output frames depicting the person wearing textually described target clothing.

公开(公告)号：US11106944B2

公开(公告)日：2021-08-31

申请号：US16557330

申请日：2019-08-30

Applicant: Adobe Inc.

Inventor： Viswanathan Swaminathan ,  Saayan Mitra ,  Han Guo

IPC: G06K9/62 ,  G06N20/00 ,  G06N3/08

Abstract: This disclosure relates to methods, non-transitory computer readable media, and systems that can initially train a machine-learning-logo classifier using synthetic training images and incrementally apply the machine-learning-logo classifier to identify logo images to replace the synthetic training images as training data. By incrementally applying the machine-learning-logo classifier to determine one or both of logo scores and positions for logos within candidate logo images, the disclosed systems can select logo images and corresponding annotations indicating positions for ground-truth logos. In some embodiments, the disclosed systems can further augment the iterative training of a machine-learning-logo classifier to include user curation and removal of incorrectly detected logos from candidate images, thereby avoiding the risk of model drift across training iterations.

公开(公告)号：US11032578B2

公开(公告)日：2021-06-08

申请号：US16020018

申请日：2018-06-27

Applicant: Adobe Inc.

Inventor： Viswanathan Swaminathan ,  Saayan Mitra ,  Akshay Malhotra

IPC: H04N7/12 ,  H04N19/94 ,  H04N19/46 ,  H04N19/91 ,  H04N21/274

Abstract: Residual vectors are compressed in a lossless compression scheme suitable for cloud DVR video content applications. Thus, a cloud DVR service provider can take many copies of the same file stored in the cloud and save storage space by compressing those copies while still maintaining their status as distinct copies, one per user. Vector quantization is used for compressing already-compressed video streams (e.g., MPEG streams). As vector quantization is a lossy compression scheme, the residual vector has to be stored to regenerate the original video stream at the decoding (playback) node. Entropy coding schemes like Arithmetic or Huffman coding can be used to compress the residual vectors. Additional strategies can be implemented to further optimize this residual compression. In some embodiments, the techniques operate to provide a 25-50% improvement in compression. Storage space is thus more efficiently used and video transmission may be faster in some cases.