公开(公告)号：US10853398B2

公开(公告)日：2020-12-01

申请号：US16189669

申请日：2018-11-13

Applicant: Adobe Inc.

Inventor： Matthew Fisher

IPC: G06F16/34 ,  G06F40/284 ,  G06F40/289 ,  G06T19/20 ,  G06F40/211 ,  G06F40/295 ,  G06F40/30 ,  G06F30/12

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for generating a three-dimensional scene based on a natural language phrase. For example, the disclosed system can analyze a natural language phrase to determine dependencies involving entities and commands in the natural language phrase. The disclosed system can then use the dependencies to generate an entity-command representation of the natural language phrase. Additionally, the disclosed system can generate a semantic scene graph for the natural language phrase from the entity-command representation to indicate contextual relationships of the entities and commands. Furthermore, the disclosed system generates the requested three-dimensional scene by using at least one scene of a plurality of available three-dimensional scenes identified using the semantic scene graph of the natural language phrase.

公开(公告)号：US10769848B1

公开(公告)日：2020-09-08

申请号：US16421729

申请日：2019-05-24

Applicant: Adobe, Inc.

Inventor： Oliver Wang ,  Vladimir Kim ,  Matthew Fisher ,  Elya Shechtman ,  Chen-Hsuan Lin ,  Bryan Russell

IPC: G06T17/20 ,  G06T15/00 ,  G06N3/08 ,  B29C64/386 ,  B33Y50/00 ,  G06T7/55

Abstract: Techniques are disclosed for 3D object reconstruction using photometric mesh representations. A decoder is pretrained to transform points sampled from 2D patches of representative objects into 3D polygonal meshes. An image frame of the object is fed into an encoder to get an initial latent code vector. For each frame and camera pair from the sequence, a polygonal mesh is rendered at the given viewpoints. The mesh is optimized by creating a virtual viewpoint, rasterized to obtain a depth map. The 3D mesh projections are aligned by projecting the coordinates corresponding to the polygonal face vertices of the rasterized mesh to both selected viewpoints. The photometric error is determined from RGB pixel intensities sampled from both frames. Gradients from the photometric error are backpropagated into the vertices of the assigned polygonal indices by relating the barycentric coordinates of each image to update the latent code vector.

公开(公告)号：US10719920B2

公开(公告)日：2020-07-21

申请号：US16188479

申请日：2018-11-13

Applicant: Adobe Inc.

Inventor： Jonathan Eisenmann ,  Zhe Lin ,  Matthew Fisher

IPC: G06T5/00 ,  G06T5/50 ,  G06T15/10 ,  G06T3/00 ,  G06T11/00

Abstract: In some embodiments, an image manipulation application receives a two-dimensional background image and projects the background image onto a sphere to generate a sphere image. Based on the sphere image, an unfilled environment map containing a hole area lacking image content can be generated. A portion of the unfilled environment map can be projected to an unfilled projection image using a map projection. The unfilled projection image contains the hole area. A hole filling model is applied to the unfilled projection image to generate a filled projection image containing image content for the hole area. A filled environment map can be generated by applying an inverse projection of the map projection on the filled projection image and by combining the unfilled environment map with the generated image content for the hole area of the environment map.