Intel Corporation

Nvidia Corporation

Tahoe Research, Ltd.

Apple Inc.

Ke Chen

Yi-Jen Chiu

Chen Wang

Ya-Ti Peng

Nikolai Smolyanskiy

Joachim Pehserl

Zhipin Deng

David Nister

Lidong Xu

Xiaoxia Cai

Ibrahim Eden

Ruchi Bhargava

Tilman Wekel

Ryan Oldja

Alexander A Popov

David Wehr

Tae-Hoon Kim

Dukhwan Kim

Jianguo Li

Sangmin Oh

Alexey Kamenev

Bongjin Jun

Bin Wang

Jim Chen Chou

Yurong Chen

Hao Pan

Haitao Guo

Charles T Loop

Krishnakanth Rapaka

Xiaohua Yang

Amir Akbarzadeh

Stan Birchfield

Francesco Iacopino

Linfeng Guo

Munehiro Nakazato

Ye Cong

Yiran Zhong

Bin Robin Wang

In various examples, perception of landmark shapes may be used for localization in autonomous systems and applications. In some embodiments, a deep neural network (DNN) is used to generate (e.g., per-point) classifications of measured 3D points (e.g., classified LiDAR points), and a representation of the shape of one or more detected landmarks is regressed from the classifications. For each of one or more classes, the classification data may be thresholded to generate a binary mask and/or dilated to generate a densified representation, and the resulting (e.g., dilated, binary) mask may be clustered into connected components that are iteratively: fitted a shape (e.g., a polynomial or Bezier spline for lane lines, a circle for top-down representations of poles or traffic lights), weighted, and merged. As such, the resulting connected components and their fitted shapes may be used to represent detected landmarks and used for localization, navigation, and/or other uses.

Landmark perception for localization in autonomous systems and applications

A deep neural network(s) (DNN) may be used to perform panoptic segmentation by performing pixel-level class and instance segmentation of a scene using a single pass of the DNN. Generally, one or more images and/or other sensor data may be stitched together, stacked, and/or combined, and fed into a DNN that includes a common trunk and several heads that predict different outputs. The DNN may include a class confidence head that predicts a confidence map representing pixels that belong to particular classes, an instance regression head that predicts object instance data for detected objects, an instance clustering head that predicts a confidence map of pixels that belong to particular instances, and/or a depth head that predicts range values. These outputs may be decoded to identify bounding shapes, class labels, instance labels, and/or range values for detected objects, and used to enable safe path planning and control of an autonomous vehicle.

Deep neural network for segmentation of road scenes and animate object instances for autonomous driving applications

In various examples, methods and systems are provided for estimating depth values for images (e.g., from a monocular sequence). Disclosed approaches may define a search space of potential pixel matches between two images using one or more depth hypothesis planes based at least on a camera pose associated with one or more cameras used to generate the images. A machine learning model(s) may use this search space to predict likelihoods of correspondence between one or more pixels in the images. The predicted likelihoods may be used to compute depth values for one or more of the images. The predicted depth values may be transmitted and used by a machine to perform one or more operations.

Predicting image warping for structure from motion using neural networks

Systems, apparatus, articles of manufacture and methods for face augmentation in video are disclosed. An example apparatus includes executable code to detect a face of a subject in the video, detect a gender of the subject based on the face, detect a skin tone of the subject based on the face, apply a first process to smooth skin on the face in the video, apply a second process to change the skin tone of the face, apply a third process to slim the face, apply a fourth process to adjust a size of eyes on the face, and apply a fifth process to remove an eye bag from the face. One or more of the first process, the second process, the third process, the fourth process, or the fifth process adjustable based on one or more of the gender or an age. The example apparatus also includes one or more processors to generate modified video with beauty effects, the beauty effects based on one or more of the first process, the second process, the third process, the fourth process, or the fifth process.

Face augmentation in video

A deep neural network(s) (DNN) may be used to detect objects from sensor data of a three dimensional (3D) environment. For example, a multi-view perception DNN may include multiple constituent DNNs or stages chained together that sequentially process different views of the 3D environment. An example DNN may include a first stage that performs class segmentation in a first view (e.g., perspective view) and a second stage that performs class segmentation and/or regresses instance geometry in a second view (e.g., top-down). The DNN outputs may be processed to generate 2D and/or 3D bounding boxes and class labels for detected objects in the 3D environment. As such, the techniques described herein may be used to detect and classify animate objects and/or parts of an environment, and these detections and classifications may be provided to an autonomous vehicle drive stack to enable safe planning and control of the autonomous vehicle.

Top-down object detection from LiDAR point clouds

Segmentation of lidar range images

In one implementation, a method of generating a bit stream encoding a video stream is performed by a device including one or more processors and non-transitory memory. The method includes decomposing a video stream into a plurality of frequency band video streams. The method includes determining a target bitrate and determining, for each frequency band video stream, a respective frequency band bit rate based on the target bit rate. The method includes encoding each of the plurality of frequency band video streams at its respective frequency band bit rate and transmitting, over a channel, each encoded frequency band video stream.

Multi-band rate control

Multi-view deep neural network for LiDAR perception

Systems, apparatus, articles, and methods are described below including operations for scalable real-time face beautification of video images.

Scalable real-time face beautification of video images

Methods and systems may provide for facial recognition of at least one input image utilizing hierarchical feature learning and pair-wise classification. Receptive field theory may be used on the input image to generate a pre-processed multi-channel image. Channels in the pre-processed image may be activated based on the amount of feature rich details within the channels. Similarly, local patches may be activated based on the discriminant features within the local patches. Features may be extracted from the local patches and the most discriminant features may be selected in order to perform feature matching on pair sets. The system may utilize patch feature pooling, pair-wise matching, and large-scale training in order to quickly and accurately perform facial recognition at a low cost for both system memory and computation.

Low-cost face recognition using Gaussian receptive field features

Systems, apparatus, articles, and methods are described below including operations for real-time face beautification features for video images.

Real-time face beautification features for video images

Techniques related to object detection using directional filtering are discussed. Such techniques may include determining directional weighted averages for pixels of an input image, generating a feature representation of the input image based on the directional weighted averages, and performing object detection by applying a multi-stage cascade classifier to the feature representation.

Object detection using binary coded images and multi-stage cascade classifiers

Growing community of inventors

Cupertino, CA, United States of America

Ke Chen