Beijing Didi Infinity Technology and Development Co., Ltd.

Ccc Information Services Inc.

Didi Research America, LLC

Ruxiao Bao

Xun Xu

Neda Hantehzadeh

Christoph Plenio

Nazanin Makkinejad

Techniques for optimizing vehicle license plate recognition in images and their decoding include training a set of convolutional neural networks (CNNs) by using images in which license plates are identified or labeled as a whole, rather than by license plate parts or key points, and rather than by the individual, segmented characters represented thereon. The trained CNNs may operate on target images of environments to localize images of license plates included therein and determine the issuing jurisdiction and/or ordered set of characters represented on detected license plates without utilizing character segmentation and/or per-character recognition techniques. As such, license plates depicted within target images are able to be detected and decoded with greater tolerances for lighting conditions, deformations or damages, occlusions, differing image resolutions, differing angles of capture, variations of other objects depicted within the images (such as dense or changing signage), etc.

Optimized vehicle license plate recognition

Methods and systems of annotation densification for semantic segmentation are disclosed herein. In one example embodiment, such a method includes obtaining image information, obtaining coarse annotation information, performing an image matting operation based upon the image information and based at least indirectly upon the coarse annotation information, and applying an already-trained Convolutional Neural Network (ConvNet) semantic segmentation model in relation to the image information. The method also includes performing a merging operation with respect to both first intermediate information generated at least indirectly by the performing of the image matting operation and second intermediate information generated at least indirectly by the applying of the ConvNet model, where the performing of the merging operation results in fine semantic segmentation annotation information, whereby an additional semantic segmentation model can be trained using that annotation information and the trained additional semantic segmentation model can be applied to generate semantic segmentation output information.

Method and system of annotation densification for semantic segmentation

Methods and systems involving convolutional neural networks as applicable for semantic segmentation, including multi-task convolutional networks employing curriculum based transfer learning, are disclosed herein. In one example embodiment, a method of semantic segmentation involving a convolutional neural network includes training and applying the convolutional neural network. The training of the convolutional neural network includes each of training a semantic segmentation decoder network of the convolutional neural network, generating first feature maps by way of an encoder network of the convolutional neural network, based at least in part upon a dataset received at the encoder network, and training an instance segmentation decoder network of the convolutional neural network based at least in part upon the first feature maps. The applying includes receiving an image, and generating each of a semantic segmentation map and an instance segmentation map in response to the receiving of the image, in a single feedforward pass.

Growing community of inventors

Mountain View, CA, United States of America

Ruxiao Bao