Beijing Kuangshi Technology Co., Ltd.

Pinhole (Beijing) Technology Co., Ltd.

Yuning Jiang

Qi Yin

Haoqiang Fan

Zhimin Cao

Erjin Zhou

Brian Li

An information processing method comprises: acquiring an original webpage; parsing the original webpage to determine a set of candidate images; detecting whether at least one candidate image in the set of candidate images contains a face; determining a target face from at least one detected face, and determining a candidate image corresponding to the target face as a target image; analyzing the target face to obtain at least one content feature of the original webpage, wherein the at least one content feature includes at least one of a face attribute feature and a personal identity feature; determining display contents matching the at least one content feature based on the at least one content feature; inserting the display contents into a target area corresponding to the target image in the original webpage in a form of a dialog box to acquire a target webpage; and outputting the target webpage.

Information processing method and information processing apparatus

A picture management method and device, a picture synchronization method and device are disclosed. The picture management method can be applied to a client device, and comprises: receiving picture tags generated by a server device performing an image analysis based on original picture features received by the server device, the original picture feature being an original picture itself or original picture parameter(s); and classifying and archiving the original pictures represented by the original picture features according to the picture tags. The picture management method can be applied to a server device, and comprises: receiving original picture features and encrypted original pictures, the original picture feature being an original picture itself or original picture parameter(s); generating pictures tag by performing an image analysis on the received original picture features, and discarding the original picture features after generating the picture tags; and classifying and archiving the encrypted original pictures represented by the original picture features according to the picture tags. An intelligent classification and archiving of pictures based on the content of the pictures can be implemented by using said methods, and backup and sharing of picture data can be implemented through the server device while ensuring security of the picture data.

Picture management method and device, picture synchronization method and device

A deep learning framework jointly optimizes the compactness and discriminative ability of face representations. The compact representation can be as compact as 32 bits and still produce highly discriminative performance. In another aspect, based on the extreme compactness, traditional face analysis tasks (e.g. gender analysis) can be effectively solved by a Look-Up-Table approach given a large-scale face data set.

Compact face representation

The present invention overcomes the limitations of the prior art by performing facial landmark localization in a coarse-to-fine manner with a cascade of neural network levels, and enforcing geometric constraints for each of the neural network levels. In one approach, the neural network levels may be implemented with deep convolutional neural network. One aspect concerns a system for localizing landmarks on face images. The system includes an input for receiving a face image, and an output for presenting landmarks identified by the system. Neural network levels are coupled in a cascade from the input to the output for the system. Each neural network level produces an estimate of landmarks. The estimate of landmarks is more refined than an estimate of landmark of a previous neural network level.

Facial landmark localization using coarse-to-fine cascaded neural networks

Face representation is a crucial step of face recognition systems. An optimal face representation should be discriminative, robust, compact, and very easy to implement. While numerous hand-crafted and learning-based representations have been proposed, considerable room for improvement is still present. A very easy-to-implement deep learning framework for face representation is presented. The framework bases on pyramid convolutional neural network (CNN). The pyramid CNN adopts a greedy-filter-and-down-sample operation, which enables the training procedure to be very fast and computation efficient. In addition, the structure of Pyramid CNN can naturally incorporate feature sharing across multi-scale face representations, increasing the discriminative ability of resulting representation.

Growing community of inventors

Beijing, China

Yuning Jiang