Abstract: Systems and methods are provided for creating shared virtual spaces using a client device. For example, the disclosed systems receive a request from a user to share one or more digital content items with other users via a shared virtual space. The disclosed systems queue a request to send data to a content management system to create the shared virtual space. In some instances, the disclosed systems can determine that a connection between a user's client device and the content management system is below a threshold connectivity, and then simulate an online experience for the user despite the poor connection.

Abstract: Systems and methods for the crowdsourced clustering of data items in accordance embodiments of the invention are disclosed. In one embodiment of the invention, a method for determining categories for a set of source data includes obtaining a set of source data, determining a plurality of subsets of the source data, where a subset of the source data includes a plurality of pieces of source data in the set of source data, generating a set of pairwise annotations for the pieces of source data in each subset of source data, clustering the set of source data into related subsets of source data based on the sets of pairwise labels for each subset of source data, and identifying a category for each related subset of source data based on the clusterings of source data and the source data metadata for the pieces of source data in the group of source data.

Abstract: Systems and methods in accordance with embodiments of the invention automatically crop images based upon the location of one or more regions of interest (ROIs) identified within the cropped image. In one embodiment, an image cropping application configures a processor to utilize at least one object detector to identify at least one region of interest within an image. A plurality of candidate cropped images within the image are selected and for each candidate cropped image, an output is generated by weighting each of the at least one region of interest contained within the candidate cropped image as a function of its position within the candidate cropped image. A crop location can then be selected based upon the candidate cropped image having the largest output.

Abstract: Embodiments of the present invention analyze and score each image associated with a group to determine representative image or images for the group. Such analysis can include detecting objects shown in the images, determining the quality of the images, and/or contextually analyzing the images as a group. In some embodiments, each image in a group (e.g., an event) of images can be analyzed by one or more image analysis modules that calculate a score for the image based on a different image characteristic. A composite image score can then be calculated based on the various image characteristic scores to identify the image or images to be used as to represent the group.

Abstract: Systems, methods, and non-transitory computer readable media are provided for maintaining local virtual states pending server-side storage across multiple devices and users and intermittent network connections. In exemplary embodiments, content added by a user to his or her account locally on a user device may be displayed, and all interactivity therewith may be facilitated, as if the content had already been created on the content management system. In content management system applications that support shared virtual spaces, changes made by the user from his or her user device to the shared virtual space (including creation of a new shared virtual space) may be displayed locally as soon as the change has been made, not waiting for the information to be transmitted to the server or its state to be made consistent with that of the mobile device.

Abstract: Embodiments are provided for ephemeral eventing. Ephemeral eventing may involve sending metadata portions associated with content items from a user device to a content management system prior to uploading the content portions of the content items. The content management system may then assign the content items to one or more events. The event assignments may then be sent back to the user device, allowing the content items to be locally displayed according to their events as if the content items had already been uploaded to the content management system.

Abstract: Systems and methods for distributed data annotation in accordance embodiments of the invention are disclosed. In one embodiment of the invention, a distributed data annotation server system includes a storage device configured to store source data, one or more annotators, annotation tasks and a processor, wherein a distributed data annotation application configures the processor to receive source data including one or more pieces of source data, select one or more annotators, create one or more annotation tasks for the selected annotators and source data, request one or more annotations for the source data using the annotation tasks, receive annotations, determine source data metadata for at least one piece of source data using the received annotations, generate annotator metadata for at least one annotator using the received annotations and the source data, and estimate the ground truth for the source data using the source data metadata and the annotator metadata.

Abstract: Systems and methods for determining annotator performance in the distributed annotation of source data in accordance embodiments of the invention are disclosed. In one embodiment of the invention, a method for clustering annotators includes obtaining a set of source data, determining a training data set representative of the set of source data, obtaining sets of annotations from a set of annotators for a portion of the training data set, for each annotator determining annotator recall metadata based on the set of annotations provided by the annotator for the training data set and determining annotator precision metadata based on the set of annotations provided by the annotator for the training data set, and grouping the annotators into annotator groups based on the annotator recall metadata and the annotator precision metadata.

Abstract: The present disclosure is directed toward systems and methods to quickly and accurately identify boundaries of a displayed document in a live camera image feed, and provide a document boundary indicator within the live camera image feed. For example, systems and methods described herein utilize different display document detection processes in parallel to generate and provide a document boundary indicator that accurately corresponds with a displayed document within a live camera image feed. Thus, a user of the mobile computing device can easily see whether the document identification system has correctly identified the displayed document within the camera viewfinder feed.

Abstract: The present disclosure is directed toward systems and methods that efficiently and effectively generate an enhanced document image of a displayed document in an image frame captured from a live image feed. For example, systems and methods described herein apply a document enhancement process to a displayed document in an image frame that result in an enhanced document image that is cropped, rectified, un-shadowed, and with dark text against a mostly white background. Additionally, systems and method described herein determine whether a stored digital content item includes a displayed document. In response to determining that a stored digital content item does include a displayed document, systems and methods described herein generate an enhanced document image of a displayed document included in the stored digital content item.

Abstract: The present disclosure is directed toward systems and methods that enable more accurate digital object classification. In particular, disclosed systems and methods address inaccuracies in digital object classification introduced by variations in classification scores. Specifically, in one or more embodiments, disclosed systems and methods generate probability functions utilizing digital test objects and transform classifications scores into normalized classification scores utilizing probability functions. Disclosed systems and methods utilize normalized classification scores to more accurately classify and identify digital objects in a variety of applications.

Abstract: Embodiments of the present invention analyze and score each image associated with a group to determine representative image or images for the group. Such analysis can include detecting objects shown in the images, determining the quality of the images, and/or contextually analyzing the images as a group. In some embodiments, each image in a group (e.g., an event) of images can be analyzed by one or more image analysis modules that calculate a score for the image based on a different image characteristic. A composite image score can then be calculated based on the various image characteristic scores to identify the image or images to be used as to represent the group.

Abstract: The present disclosure is directed toward systems and methods that enable more accurate digital object classification. In particular, disclosed systems and methods address inaccuracies in digital object classification introduced by variations in classification scores. Specifically, in one or more embodiments, disclosed systems and methods generate probability functions utilizing digital test objects and transform classifications scores into normalized classification scores utilizing probability functions. Disclosed systems and methods utilize normalized classification scores to more accurately classify and identify digital objects in a variety of applications.

Abstract: Recommendations for shareable links to content items stored in an online content management service. In one aspect, a recommendation is sent to a first user to create and share a shareable link to a content item with a second user. The content item to share via a shareable link is identified as one that is currently not shared by the first user and that has one or more characteristics in common with a “popular” content item as determined by monitoring the popular content item for a threshold number of requests to access the popular content item via a corresponding shareable link to the popular content item.

Abstract: Monitored shareable links to content items in an online content management service. In one aspect, based on determining that a first view metric associated with a shareable link to a content item is above a first activity threshold associated with the shareable link, a first action is taken on the shareable link. The first action can be allowing or denying requested access to the content item via the shareable link or sending a notification to a user that requested creation of the shareable link regarding user activity on the shareable link. Further, based on determining that a second view metric associated with the shareable link is above a second activity threshold associated with the shareable link, a second action is taken on the shareable link. The second action can be like the first action.

Abstract: Systems and methods for the annotation of source data using confidence labels in accordance embodiments of the invention are disclosed. In one embodiment of the invention, a method for determining confidence labels for crowdsourced annotations includes obtaining a set of source data, obtaining a set of training data representative of the set of source data, determining the ground truth for each piece of training data, obtaining a set of training data annotations including a confidence label, measuring annotator accuracy data for at least one piece of training data, and automatically generating a set of confidence labels for the set of unlabeled data based on the measured annotator accuracy data and the set of annotator labels used.

Abstract: The present disclosure is directed toward systems and methods that enable more accurate digital object classification. In particular, disclosed systems and methods address inaccuracies in digital object classification introduced by variations in classification scores. Specifically, in one or more embodiments, disclosed systems and methods generate probability functions utilizing digital test objects and transform classifications scores into normalized classification scores utilizing probability functions. Disclosed systems and methods utilize normalized classification scores to more accurately classify and identify digital objects in a variety of applications.

Abstract: A content item on a content management system can be shared using a generated sharing link. The sharing link can be a custom network address, such as a uniform resource locator (URL), which allows the content item to be accessed without authentication. The sharing index and a content path of the content items can be listed in a sharing index used to identify the content upon a request initiated by the sharing link. The content management system can generate a content link to a content item upon receiving a sharing input from a user indicating the user wants to share a content item. Alternatively, in some embodiments, sharing links can be pre-generated by the content management system and stored on the client device. The sharing link can stored directly to a data buffer on the client device, such as a clipboard, where it can be accessed by the user.

Abstract: Shareable links can be created to share content items and information pertaining to activity on those shareable links can be monitored and stored in an online content management service. Based on this activity information, actions can be executed. These actions can include sending notifications regarding link activity, disabling shareable links, and/or updating activity thresholds related to the shareable links. Using the activity information, popular shared content items can be identified and recommendations for sharing unshared content items can be provided to the user. Additionally, advertisements can be tailored to the relative popularity of the shared content items.

Abstract: Shareable links can be created to share content items and information pertaining to activity on those shareable links can be monitored and stored in an online content management service. Based on this activity information, predetermined actions can be executed. These actions can include sending notifications regarding link activity, disabling shareable links, and/or updating sharing limitations related to the shareable links. Using the activity information, popular shared content items can be identified and recommendations for sharing unshared content items can be provided to the user. Additionally, advertisements can be tailored to the relative popularity of the shared content items.