Toyota Research Institute, Inc.

Toyota Jidosha Kabushiki Kaisha

Toyota Technological Institute at Chicago

Leland Stanford Junior University

Massachusetts Institute of Technology

Woven Planet North America, Inc.

Vitor Guizilini

Adrien David Gaidon

Rares A Ambrus

Sudeep Pillai

Jie Li

Igor Vasiljevic

Sergey Zakharov

Dennis Park

Jiexiong Tang

Gregory Shakhnarovich

Hanme Kim

Kuan-Hui Lee

Charles Christopher Ochoa

Jiading Fang

Matthew R Walter

Victor Vaquero Gomez

Arjun Bhargava

Leonidas John Guibas

Or Litany

Rui Hou

Alexander Alspach

Kun-Hsin Chen

Naveen Suresh Kuppuswamy

Chao Fang

Wadim Kehl

Andrew M Beaulieu

Fredo Durand

Jia-En M Pan

Xipeng Wang

Dian Chen

Joshua Tenenbaum

Jiajun Wu

Jeff Walls, Sr

Greg Shakhnarovich

Yue Wang

Xiangru Huang

Justin Solomon

Vincent Sitzmann

A method of generating additional supervision data to improve learning of a geometrically-consistent latent scene representation with a geometric scene representation architecture is provided. The method includes receiving, with a computing device, a latent scene representation encoding a pointcloud from images of a scene captured by a plurality of cameras each with known intrinsics and poses, generating a virtual camera having a viewpoint different from viewpoints of the plurality of cameras, projecting information from the pointcloud onto the viewpoint of the virtual camera, and decoding the latent scene representation based on the virtual camera thereby generating an RGB image and depth map corresponding to the viewpoint of the virtual camera for implementation as additional supervision data.

Geometric 3D augmentations for transformer architectures

A method estimating a depth of an environment includes generating, via a cross-attention model, a cross-attention cost volume based on a current image of the environment and a previous image of the environment in a sequence of images. The method also includes generating, via the cross-attention model, a depth estimate of the current image based on the cross-attention cost volume, the cross-attention model having been trained using a photometric loss associated with a single-frame depth estimation model. The method further includes controlling an action of the vehicle based on the depth estimate.

Photometric masks for self-supervised depth learning

Systems and methods for enhanced computer vision capabilities, particularly including depth synthesis, which may be applicable to autonomous vehicle operation are described. A vehicle may be equipped with a geometric scene representation (GSR) architecture for synthesizing depth views at arbitrary viewpoints. The GSR architecture synthesizes depth views enable advanced functions, including depth interpolation and depth extrapolation. The GSR architecture implements functions (i.e., depth interpolation, depth extrapolation) that are useful for various computer vision applications for autonomous vehicles, such as predicting depth maps from unseen locations. For example, a vehicle includes a processor device synthesizing depth views at multiple viewpoints, where the multiple viewpoints are from image data of a surrounding environment for the vehicle. Further, the vehicle can have a controller device that receives depth views from the processor device and performs autonomous operations in response to analysis of the depth views.

Systems and methods for depth synthesis with transformer architectures

A method of representation learning for object detection from unlabeled point cloud sequences is described. The method includes detecting moving object traces from temporally-ordered, unlabeled point cloud sequences. The method also includes extracting a set of moving objects based on the moving object traces detected from the sequence of temporally-ordered, unlabeled point cloud sequences. The method further includes classifying the set of moving objects extracted from on the moving object traces detected from the sequence of temporally-ordered, unlabeled point cloud sequences. The method also includes estimating 3D bounding boxes for the set of moving objects based on the classifying of the set of moving objects.

Representation learning for object detection from unlabeled point cloud sequences

A method for improving 3D object detection via object-level augmentations is described. The method includes recognizing, using an image recognition model of a differentiable data generation pipeline, an object in an image of a scene. The method also includes generating, using a 3D reconstruction model, a 3D reconstruction of the scene from the image including the recognized object. The method further includes manipulating, using an object level augmentation model, a random property of the object by a random magnitude at an object level to determine a set of properties and a set of magnitudes of an object manipulation that maximizes a loss function of the image recognition model. The method also includes training a downstream task network based on a set of training data generated based on the set of properties and the set of magnitudes of the object manipulation, such that the loss function is minimized.

Adversarial object-aware neural scene rendering for 3D object detection

System, methods, and other embodiments described herein relate to an improved approach to training a depth model for monocular depth estimation by warping depth features prior to decoding. In one embodiment, a method includes encoding, using an encoder of a depth model, a source image into depth features of a scene depicted by the source image. The method includes warping the depth features into warped features of a target frame of a target image associated with the source image. The method includes decoding, using a decoder of the depth model, the warped features into a depth map. The method includes training the depth model according to a loss derived from the depth map.

Warping depth features for depth estimation

Systems and methods are provided for depth estimation from monocular images using a depth model with sparse range sensor data and uncertainty in the range sensor as inputs thereto. According to some embodiments, the methods and systems comprise receiving an image captured by an image sensor, where the image represents a scene of an environment. The method and systems also comprise deriving a point cloud representative of the scene of the environment from range sensor data, and deriving range sensor uncertainty from the range sensor data. Then a depth map can be derived for the image based on the point cloud and the range sensor uncertainty as one or more inputs into a depth model.

Depth estimation with sparse range sensor depth and uncertainty projection

System, methods, and other embodiments described herein relate to an improved approach to training a depth model to derive depth estimates from monocular images using histograms to assess photometric losses. In one embodiment, a method includes determining loss values according to a photometric loss function. The loss values are associated with a depth map derived from an input image that is a monocular image. The method includes generating histograms for the loss values corresponding to different regions of a target image. The method includes, responsive to identifying erroneous values of the loss values, masking the erroneous values to avoid considering the erroneous values during training of the depth model.

Using histograms for self-supervised depth estimation

System, methods, and other embodiments described herein relate to an improved approach to training a depth model to derive depth estimates from monocular images using cost volumes. In one embodiment, a method includes predicting, using a depth model, depth values from at least one input image that is a monocular image. The method includes generating a cost volume by sampling the depth values corresponding to bins of the cost volume. The method includes determining loss values for the bins of the cost volume. The method includes training the depth model according to the loss values of the cost volume.

Photometric cost volumes for self-supervised depth estimation

Systems and methods are provided for training a depth model to recover scale factor for self-supervised depth estimation in monocular images. Examples include deriving a depth map for an image based on a depth model. The depth map comprises depth values for pixels of the image. A first scale for the image can be estimated based on the depth values, and depth data captured by a range sensor can be received. The depth data comprises a point cloud comprising depth measures. A second scale for the point cloud can be determined based on the depth measures and a scale factor can be determined based the second scale and the first scale. The depth model can be updated based on the scale factor, wherein the depth model generates metrically accurate depth estimates based on the scale factor.

Scale-aware self-supervised monocular depth with sparse radar supervision

Systems and methods for self-supervised learning for visual odometry using camera images, may include: estimating correspondences between keypoints of a target camera image and keypoints of a context camera image; based on the keypoint correspondences, lifting a set of 2D keypoints to 3D, using a neural camera model; and projecting the 3D keypoints into the context camera image using the neural camera model. Some embodiments may use the neural camera model to achieve the lifting and projecting of keypoints without a known or calibrated camera model.

Systems and methods for generic visual odometry using learned features via neural camera models

A method for 3D object detection is described. The method includes concurrently training a monocular depth network and a 3D object detection network. The method also includes predicting, using a trained monocular depth network, a monocular depth map of a monocular image of a video stream. The method further includes inferring a 3D point cloud of a 3D object within the monocular image according to the predicted monocular depth map. The method also includes predicting 3D bounding boxes from a selection of 3D points from the 3D point cloud of the 3D object based on a selection regression loss.

Monocular object detection via end-to-end differentiable pipeline

System, methods, and other embodiments described herein relate to training a depth model for monocular depth estimation using photometric loss masks derived from motion estimates of dynamic objects. In one embodiment, a method includes generating depth maps from images of an environment. The method includes determining motion of points within the depth maps. The method includes associating the points between the depth maps to identify an object according to a correlation of the motion for a first cluster of the points with a second cluster of the points. The method includes providing the depth maps and the object as an electronic output.

Self-supervised monocular depth estimation via rigid-motion embeddings

A method for controlling an agent to navigate through an environment includes generating a depth map associated with a monocular image of the environment. The method also includes generating a group of surface normal. Each surface normal of the group of surface normals is associated with a respective polygon of a group of polygons associated with the depth map. The method further includes identifying one or more ground planes in the depth map based on the group of surface normal. The method further includes controlling the agent to navigate through the environment based on identifying the one or more ground planes.

System and method for self-supervised monocular ground-plane extraction

A method for controlling a vehicle in an environment includes generating, via a cross-attention model, a cross-attention cost volume based on a current image of the environment and a previous image of the environment in a sequence of images. The method also includes generating combined features by combining cost volume features of the cross-attention cost volume with single-frame features associated with the current image. The single-frame features may be generated via a single-frame encoding model. The method further includes generating a depth estimate of the current image based on the combined features. The method still further includes controlling an action of the vehicle based on the depth estimate.

Growing community of inventors

Santa Clara, CA, United States of America

Vitor Guizilini