Entity

Time filter

Source Type

Seattle, WA, United States

Amazon.com, Inc. is an American electronic commerce company with headquarters in Seattle, Washington. It is the largest Internet-based retailer in the United States. Amazon.com started as an online bookstore, but soon diversified, selling DVDs, VHSs, CDs, video and MP3 downloads/streaming, software, video games, electronics, apparel, furniture, food, toys, and jewelry. The company also produces consumer electronics—notably, Amazon Kindle e-book readers, Fire tablets, Fire TV and Fire Phone — and is a major provider of cloud computing services.Amazon has separate retail websites for United States, United Kingdom & Ireland, France, Canada, Germany, The Netherlands, Italy, Spain, Australia, Brazil, Japan, China, India and Mexico. Amazon India will soon start offering music, movie and video streaming services in India. Amazon also offers international shipping to certain other countries for some of its products. In 2011, it had professed an intention to launch its websites in Poland and Sweden. Wikipedia.


Ren X.,Amazon | Ramanan D.,University of California at Irvine
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition | Year: 2013

Object detection has seen huge progress in recent years, much thanks to the heavily-engineered Histograms of Oriented Gradients (HOG) features. Can we go beyond gradients and do better than HOG? We provide an affirmative answer by proposing and investigating a sparse representation for object detection, Histograms of Sparse Codes (HSC). We compute sparse codes with dictionaries learned from data using K-SVD, and aggregate per-pixel sparse codes to form local histograms. We intentionally keep true to the sliding window framework (with mixtures and parts) and only change the underlying features. To keep training (and testing) efficient, we apply dimension reduction by computing SVD on learned models, and adopt supervised training where latent positions of roots and parts are given externally e.g. from a HOG-based detector. By learning and using local representations that are much more expressive than gradients, we demonstrate large improvements over the state of the art on the PASCAL benchmark for both root-only and part-based models. © 2013 IEEE.


A policy is incorporated into a first set of policies at least in part by generating a second set of policies corresponding to the policy. An index of the first set of policies is generated based at least in part on a policy element of a normal form. Based at least in part on the index, a subset of the first set of policies that is relevant to at least one of a plurality of policy enforcement components is identified and provided to at least one of the plurality of policy enforcement components of a virtual resource provider identified as relevant. A request subject to the policy is received, and the policy is enforced at least in part by evaluating the request with respect to the subset of the first set of policies.


Patent
Amazon | Date: 2016-01-08

Techniques are described for facilitating interactions with device driver modules. In at least some situations, the techniques include managing interactions between device driver modules and other programs or hardware devices so as to minimize disruptions related to the device driver modules, including when changes to existing device driver modules are made. Such device driver module changes may have various forms and may occur for various reasons, including to install new versions of device driver modules or otherwise upgrade existing device driver modules. Furthermore, the interactions with device driver modules may be managed in various manners, including to allow changes to occur to a device driver module while that device driver module is in use on a computing system, but without causing other programs on the computing system to be restarted or to lose existing connections to the device driver module being changed.


Patent
Amazon | Date: 2016-01-11

A set of techniques is described for enabling a virtual machine based transcoding system. The system enables any transcoding provider to make their transcoding service available to other users over a network. The system can automate the deployment, execution and delivery of the transcoding service on behalf of the transcoding provider and enable other users to use the transcoding services to transcode content. The system receives a virtual machine image, transfers the image to a location where the media content is stored and creates a virtual private network of resources that will perform the transcoding of the media content. The virtual private network may be firewalled or otherwise restricted from opening connections with external clients when transcoding the content in order to prevent malicious use of the media content.


Systems and methods for providing object versioning in a storage system may support the logical deletion of stored objects. In response to a delete operation specifying both a user key and a version identifier, the storage system may permanently delete the specified version of an object having the specified key. In response to a delete operation specifying a user key, but not a version identifier, the storage system may create a delete marker object that does not contain object data, and may generate a new version identifier for the delete marker. The delete marker may be stored as the latest object version of the user key, and may be addressable in the storage system using a composite key comprising the user key and the new version identifier. Subsequent attempts to retrieve the user key without specifying a version identifier may return an error, although the object was not actually deleted.

Discover hidden collaborations