Carnegie Mellon University

Fangyi Chen

Marios Savvides

Chenchen Zhu

Uzair Ahmed

Ran Tao

Zhiqiang Shen

Han Zhang

Shayeree Sarkar

Disclosed herein is a system and method for improving the accuracy of an object detector when trained with a dataset having a significant number of missing annotations. The method uses a novel Background Recalibration Loss (BRL) which adjusts the gradient direction according to its own activation to reduce the adverse effect of error signals by replacing the negative branch of the focal loss with a mirror of the positive branch when the activation is below a confusion threshold.

System and method for solving missing annotation object detection

Disclosed herein is an improved few-shot detector which utilizes a dynamic semantic network which takes as input a language feature and generates trainable parameters for a visual network. The visual network takes a visual feature as input and generates a classification and localization of an object.

System and method for improved few-shot object detection using a dynamic semantic network

Disclosed herein is an improved method for identifying images containing objects-of-interest from a large set of images. The method comprises mixing two or more of the images to create a grouped image and exposing the grouped image to an object detector trained on grouped images to make an initial determination that the grouped image was formed from at least one image containing an object-of-interest. The images which formed the grouped image are then exposed to regular object detectors to determine a classification of the object-of-interest.

Fast object search based on the cocktail party effect

Disclosed herein is an improved few-shot detector which utilizes semantic relation reasoning to learn novel objects from both visual information and the semantic relation of base class objects Specifically, a semantic space is constructed using word embeddings. Guided by the word embeddings of the classes, the detector is trained to project the objects from the visual space to the semantic space and to align their image representations with the corresponding class embeddings.

Few-shot object detection using semantic relation reasoning

Disclosed herein is an improvement to prior art feature pyramids for general object detection that inserts a simple norm calibration (NC) operation between the feature pyramids and detection head to alleviate and balance the norm bias caused by feature pyramid network (FPN) and which leverages an enhanced multi-feature selective strategy (MS) during training to assign the ground-truth to one or more levels of the feature pyramid.

Feature pyramids for object detection

Disclosed herein is a system and method of identifying new products on a retail shelf using a feature extractor trained to extract features from images of products on the shelf and output identifying information regarding the product in the product image. The extracted features are compared to extracted features in a product library and a best-fit is obtained. A new product is identified if the distance between the features of the product on the shelf and the features of the best-fit product from the product library are above a predetermined threshold.

Growing community of inventors

Pittsburgh, PA, United States of America

Fangyi Chen