Shenzhen Malong Technologies Co., Ltd.

Other

Matthew Robert Scott

Weilin Huang

Sheng Guo

Dinglong Huang

Haozhi Zhang

Limin Wang

Le Yin

Shiwen Zhang

Kai Fu

Yu Gao

Dengke Dong

Xiaojun Hu

Tiangang Zhang

Xintong Han

Chenfan Zhuang

Luchen Liu

This disclosure includes technologies for action recognition in general. The disclosed system may automatically detect various types of actions in a video, including reportable actions that cause shrinkage in a practical application for loss prevention in the retail industry. Further, appropriate responses may be invoked if a reportable action is recognized. In some embodiments, a three-branch architecture may be used in a machine learning model for action and/or activity recognition. The three-branch architecture may include a main branch for action recognition, an auxiliary branch for learning/identifying an actor (e.g., human parsing) related to an action, and an auxiliary branch for learning/identifying a scene related to an action. In this three-branch architecture, the knowledge of the actor and the scene may be integrated in two different levels for action and/or activity recognition.

Retail inventory shrinkage reduction via action recognition

Methods and systems are provided for generating a multi-label classification system. The multi-label classification system can use a multi-label classification neural network system to identify one or more labels for an image. The multi-label classification system can explicitly take into account the relationship between classes in identifying labels. A relevance sub-network of the multi-label classification neural network system can capture relevance information between the classes. Such a relevance sub-network can decouple independence between classes to focus learning on relevance between the classes.

Decoupling category-wise independence and relevance with self-attention for multi-label image classification

Methods and systems for training machine vision models (MVMs) with 'noisy' training datasets are described. A noisy set of images is received, where labels for some of the images are 'noisy' and/or incorrect. A progressively-sequenced learning curriculum is designed for the noisy dataset, where the images that are easiest to learn machine-vision knowledge from are sequenced near the beginning of the curriculum and images that are harder to learn machine-vision knowledge from are sequenced later in the curriculum. An MVM is trained via providing the sequenced curriculum to a supervised learning method, so that the MVM learns from the easiest examples first and the harder training examples later, i.e., the MVM progressively accumulates knowledge from simplest to most complex. To sequence the curriculum, the training images are embedded in a feature space and the &#x201c;complexity&#x201d; of each image is determined via density distributions and clusters in the feature space.

Complexity-based progressive training for machine vision models

Aspects of this disclosure include technologies for appearance-flow-based image generation. In applications for pose-guided person image generation or virtual try-on, the disclosed system can model the appearance flow between source and target clothing regions. Further, a cascaded appearance flow estimation network is used to progressively refine the appearance flow. The resulting appearance flow can properly encode the geometric changes between the source and the target for image generation.

Appearance-flow-based image generation

Aspects of this disclosure include technologies to detect unpackaged, unlabeled, or mislabeled products based on product images. Leveraging from improved machine learning techniques, the disclosed technical solution can reduce a full product space for product search to a partial product space. Accordingly, a limited number of product candidates in the partial product space may be visually presented for user evaluation. Sometimes, a product candidate is to be comparatively presented with a live image of the product via a uniquely designed graphic user interface, which further improves the confidence of the user and the accuracy of the underlying transaction.

Compressed network for product recognition

Aspects of this disclosure include technologies for detecting mislabeled products. In one embodiment, the disclosed system will capture an image of a product when the MRL of the product is scanned or being scanned. After recognizing the product in the image, the size of the area containing the product may be calculated. Subsequently, the disclosed system can determine whether the MRL mismatches the product in the image if this size of the area containing the product does not match the standard size associated with the MRL.

Mislabeled product detection

This disclosure includes technologies for video recognition in general. The disclosed system can automatically detect various types of actions in a video, including reportable actions that cause shrinkage in a practical application for loss prevention in the retail industry. The temporal evolution of spatio-temporal features in the video are used for action recognition. Such features may be learned via a 4D convolutional operation, which is adapted to model low-level features based on a residual 4D block. Further, appropriate responses may be invoked if a reportable action is recognized.

4D convolutional neural networks for video recognition

Provided are a method and apparatus for identifying a product. The method includes: acquiring an image of the product; performing multilevel detection on the image to determine a label region of the product, specifically, an image region corresponding to a previous level of detection is greater than an image region corresponding to a following level of detection; and identifying information in the label region to determine information of the product. A product can be identified automatically by acquiring the image of the product and performing multilevel detection on the image, thereby efficiency is improved, a large number of products can be handled, and cost is reduced.

Method and apparatus for identifying a product

Provided are a self-service method and device. The method includes: acquiring by a camera video data associated with a shopping cart at a distance from a shelf less than a preset value; determining whether an article is being placed into the shopping cart, based on identification of the video data; if yes, acquiring an ID of the shopping cart, determining the placed article and associating the ID with the article; displaying a total sum of all articles associated with the ID to the user when detecting the shopping cart corresponding to the ID is located in a preset payment area in the shopping mall; and executing a permitting procedure for allowing the shopping cart corresponding to the ID to pass upon reception of a message fed back from the payment area and indicating that the total sum corresponding to the ID has been cleared.

Growing community of inventors

Shenzhen, China

Matthew Robert Scott