Abstract: It is provided a method for detecting objects in an image, the method being performed by a mobile device. The method comprises: obtaining an image to process; performing tracking to find any tracked objects in the image, wherein any tracked objects have been identified in a previous image; performing category object detecting of the image, resulting in a respective category label for each detected object; and sending a request to a server to perform specific object detecting of at least one object detected in the category object detecting, wherein the specific object detecting is a more granular object detecting than the category object detecting.

Abstract: The present disclosure provides system and methods of bitrate adaptation for edge-assisted simultaneous localization and mapping (SLAM) where the bitrate of image data sent from a remote device to an edge device or server, can be dynamically adapted based on network availability. A model can be constructed offline that determines an optimal bitrate for the image data sent to a network node of a wireless communication system to improve and/or maximize localization performance based on a set of parameters including network latency, whether a map of an environment is present, and the bitrate of the map. The network latency can be determined based on another model that determines the latency as a function of the image data bitrate and network availability. Once the bitrate is determined, the image data can be compressed such that the compressed image data is transmitted to the network node at the determined bitrate.

Abstract: A computer-implemented method is provided performed by a computing device for determining a measure of an object detection failure based on an awareness of network quality and assistance from a distributed device. The method includes determining a first result of a first object detector of the computing device; and determining a local state of the computing device that includes a first metric of a probability of a failure of the first object detector. The method further includes receiving, from the distributed device, a second result of a second object detector; and determining a global state including a value based on a comparison of the first and second result; obtaining a third metric of a quality of the network; and outputting the measure of the object detection failure based on a merge of the local state, the global state, and the third metric.

Abstract: The present disclosure provides an adaptive compression system performed by a mobile device for image data to be sent to an edge device that takes into consideration the presence and/or bitrate of any maps of the environment. In an embodiment, the mobile device can determine whether image data associated with an image of an environment corresponds to at least a portion of a map of the environment and determine a first bitrate of the map. The mobile device can, in response to determining that the image data matches at least a portion of the map of the environment, encode the image data, where a second bitrate of the encoded image data is selected based on whether there is a corresponding map and on the first bitrate of the map. The mobile device can then transmit the encoded image data to a network node.

Abstract: There is provided techniques for determining whether or not to include an image in a 3D model of a scene. A method is performed by an image processing device. The method is repeatedly performed for images obtained from an image scan of the scene that have not already been included in the 3D model. The method for each iteration k=1:K comprises obtaining a pose as determined for an image selected for iteration k and data used to compute the pose for the selected image, and determining, depending on whether a value for the image as computed as a function of the pose and the data for the selected image fulfils a quality criterion or not, whether, and when, to include the selected image in the 3D model. The selected image is included in the 3D model for iteration k when the value fulfils the quality criterion. The selected image is not included in the 3D model for iteration k when the value fails to fulfil the quality criterion.

Abstract: Systems and methods are disclosed for joint bitrate adaptation and feature extraction for server-assisted localization of mobile devices. In one embodiment, a method performed by a mobile device comprises obtaining an image of an environment in which the mobile device is located, determining a desired bitrate for compression of the image, and determining whether one or more criteria for transmission of the image at the desired bitrate are satisfied. The method further comprises, if the criteria are not satisfied, performing a feature extraction procedure that detects features in the image and provides feature information comprising information that describes the detected features and information that indicates locations of the features within the image and transmitting the feature information to the associated server. The method further comprises, if any of the criteria are satisfied, compressing the image at the desired bitrate and transmitting the compressed image to the associated server.

Abstract: A method for processing images. The method includes obtaining a first image and a second image. The method also includes dividing the first image into N regions. The method also includes, for a first region of the first image, defining a corresponding first region of the second image, and, for a second region of the first image, defining a corresponding second region of the second image. The method also includes detecting a first feature in the first region of the first image and detecting a second feature in the second region of the first image. The method also includes searching the second image for a feature matching the first feature, wherein the searching of the second image for the feature matching the first feature is limited to searching only the corresponding first region of the second image.

Abstract: It is provided a method for performing 3D reconstruction. The method includes: obtaining sensor data; determining a pose estimate; estimating a pose error; comparing the pose error against an error threshold; performing a device 3D reconstruction when the pose error is determined to be smaller than the error threshold, resulting in updates to a device 3D model; sending a 3D reconstruction request to the server to perform a central 3D reconstruction, when the pose error is determined to be greater than the error threshold, wherein the 3D reconstruction request includes data based on the sensor data; receiving a result of a central 3D reconstruction from the server; and performing a 3D model fusion of a device 3D model in the mobile device and the result of the central 3D reconstruction, wherein the device 3D model, at least partly, is a result of previous device 3D reconstruction.