Baidu USA LLC

Baidu.com Times Technology (beijing) Co. Ltd.

Miao Liao

Sibo Zhang

Ruigang Yang

Feixiang Lu

Liangjun Zhang

Peng Wang

Dingfu Zhou

Jiahong Yuan

Presented herein are novel approaches to synthesize video of the speech from text. In a training phase, embodiments build a phoneme-pose dictionary and train a generative neural network model using a generative adversarial network (GAN) to generate video from interpolated phoneme poses. In deployment, the trained generative neural network in conjunction with the phoneme-pose dictionary convert an input text into a video of a person speaking the words of the input text. Compared to audio-driven video generation approaches, the embodiments herein have a number of advantages: 1) they only need a fraction of the training data used by an audio-driven approach; 2) they are more flexible and not subject to vulnerability due to speaker variation; and 3) they significantly reduce the preprocessing, training, and inference times.

Text-driven video synthesis with phonetic dictionary

Presented herein are novel embodiments for converting a given speech audio or text into a photo-realistic speaking video of a person with synchronized, realistic, and expressive body dynamics. In one or more embodiments, 3D skeleton movements are generated from the audio sequence using a recurrent neural network, and an output video is synthesized via a conditional generative adversarial network. To make movements realistic and expressive, the knowledge of an articulated 3D human skeleton and a learned dictionary of personal speech iconic gestures may be embedded into the generation process in both learning and testing pipelines. The former prevents the generation of unreasonable body distortion, while the later helps the model quickly learn meaningful body movement with a few videos. To produce photo-realistic and high-resolution video with motion details, a part-attention mechanism is inserted in the conditional GAN, where each detailed part is automatically zoomed in to have their own discriminators.

Personalized speech-to-video with three-dimensional (3D) skeleton regularization and expressive body poses

Systems and methods of video inpainting for autonomous driving are disclosed. For example, the method stitches a multiplicity of depth frames into a 3D map, where one or more objects in the depth frames have previously been removed. The method further projects the 3D map onto a first image frame to generate a corresponding depth map, where the first image frame includes a target inpainting region. For each target pixel within the target inpainting region of the first image frame, based on the corresponding depth map, the method further maps the target pixel within the target inpainting region of the first image frame to a candidate pixel in a second image frame. The method further determines a candidate color to fill the target pixel. The method further performs Poisson image editing on the first image frame to achieve color consistency at a boundary and between inside and outside of the target inpainting region of the first image frame. For each pixel in the target inpainting region of the first image frame, the method further traces the pixel into neighboring frames and replacing an original color of the pixel with an average of colors sampled from the neighboring frames.

Depth-guided video inpainting for autonomous driving

Growing community of inventors

Sunnyvale, CA, United States of America

Miao Liao