公开(公告)号：US20240331279A1

公开(公告)日：2024-10-03

申请号：US18194793

申请日：2023-04-03

Applicant: Honda Motor Co., Ltd.

Inventor： Kunal NANDANWAR ,  Nawid JAMALI ,  Hifza JAVED

IPC: G06T17/00 ,  G06V40/16

CPC classification number: G06T17/00 ,  G06V40/168

Abstract: A system and method for completing three dimensional face reconstruction that includes receiving image data associated with multiple two dimensional non-frontal face images. The system and method also includes analyzing the image data and extracting two dimensional facial features. The system and method additionally includes constructing sparse three dimensional facial feature point clouds based on the two dimensional facial features. The system and method further includes inputting the sparse three dimensional facial feature point clouds into an encoder-decoder architecture to generate a three dimensional facial feature point cloud of complete facial features.

公开(公告)号：US20240326256A1

公开(公告)日：2024-10-03

申请号：US18194856

申请日：2023-04-03

Applicant: Honda Motor Co., Ltd.

Inventor： Hifza JAVED ,  Nawid JAMALI

IPC: B25J11/00 ,  G06N3/091

CPC classification number: B25J11/0015 ,  B25J11/001 ,  G06N3/091

Abstract: According to one aspect, behavior generation for situationally-aware social robots may include generating a synthesized behavior based on a latent space representation indicative of a situational context including a situationally-aware social robot and one or more individuals, performing behavior retargeting on the synthesized behavior based on a real behavior from an initial latent space input to generate a retargeted behavior, and performing the retargeted behavior. The generating the synthesized behavior and the performing behavior retargeting may be implemented based on a generative adversarial network (GAN) based network. The GAN based network may be trained using a multimodal human behavioral dataset.