Abstract: The invention relates to a system for detecting objects in a digital image. The system comprises a neural network which is configured to generate candidate windows indicating object locations, and to generate for each candidate window a score representing the confidence of detection. Generating the scores comprises: generating a latent representation for each candidate window, updating the latent representation of each candidate window based on the latent representation of neighboring candidate windows, and generating the score for each candidate window based on its updated latent representation The invention further relates to a system for rescoring object detections in a digital image and to methods of detecting objects and rescoring objects.

Abstract: A method and system for training a model for image generation. The model includes a hybrid variational auto-encoder (VAE)—generative adversarial network (GAN) framework. The method includes the steps of: multiple input of an input image into the VAE which outputs in response multiple distinct output image samples, determining the best of the multiple output image samples as a best-of-many sample, the best-of-many sample having the minimum reconstruction cost, and training the model based on a predefined training objective, the predefined training objective integrating the best-of-many sample reconstruction cost and a GAN-based synthetic likelihood term.

Abstract: A method for training a model to be used for processing images, wherein the model comprises: —a first portion (101) configured to receive images as input and configured to output a feature map, —a second portion (102) configured to receive the feature map outputted by the first portion as input and configured to output a semantic segmentation, the method comprising: —training a generator (201) so that the generator is configured to generate a feature map configured to be used as input to the second portion, —generating a plurality of feature maps using the generator, —training the second portion using the feature maps generated by the generator.

Abstract: The invention relates to a system for detecting objects in a digital image. The system comprises a neural network which is configured to generate candidate windows indicating object locations, and to generate for each candidate window a score representing the confidence of detection. Generating the scores comprises: generating a latent representation for each candidate window, updating the latent representation of each candidate window based on the latent representation of neighboring candidate windows, and generating the score for each candidate window based on its updated latent representation The invention further relates to a system for rescoring object detections in a digital image and to methods of detecting objects and rescoring objects.

Abstract: A system for tracking objects in a temporal sequence of digital images is configured to: detect potential objects in the images, the detected potential objects being indicated as nodes, identify pairs of neighboring nodes, such that for each pair the nodes of said pair potentially represent an identical object and their spatial and/or temporal relationship with each other is within a predetermined range, connect each pair of neighboring nodes with a first type edge, identify at least one supplementary pair of distant nodes whose spatial and/or temporal relationship with each other exceeds the predetermined range, connect the pair of distant nodes with a supplementary second type edge, each of the first and second type edges being assigned a cost value, and determine a track of an object in the temporal sequence of digital images based on a set of connected first type edges and at least one second type edge.

Abstract: A system for tracking objects in a temporal sequence of digital images is configured to: detect potential objects in the images, the detected potential objects being indicated as nodes, identify pairs of neighboring nodes, such that for each pair the nodes of said pair potentially represent an identical object and their spatial and/or temporal relationship with each other is within a predetermined range, connect each pair of neighboring nodes with a first type edge, identify at least one supplementary pair of distant nodes whose spatial and/or temporal relationship with each other exceeds the predetermined range, connect the pair of distant nodes with a supplementary second type edge, each of the first and second type edges being assigned a cost value, and determine a track of an object in the temporal sequence of digital images based on a set of connected first type edges and at least one second type edge.

Abstract: The invention relates to a system for detecting objects in a digital image. The system comprises a neural network which is configured to generate candidate windows indicating object locations, and to generate for each candidate window a score representing the confidence of detection. Generating the scores comprises: generating a latent representation for each candidate window, updating the latent representation of each candidate window based on the latent representation of neighboring candidate windows, and generating the score for each candidate window based on its updated latent representation The invention further relates to a system for rescoring object detections in a digital image and to methods of detecting objects and rescoring objects.

Abstract: An image processor (10) has a window selector for choosing a detection window within the image, and a self similarity computation part (40) for determining self-similarity information for a group of the pixels in any part of the detection window, to represent an amount of self-similarity of that group to other groups in any other part of the detector window, and for repeating the determination for groups in all parts of the detection window, to generate a global self similarity descriptor for the detection window. A classifier (50) is used for classifying whether an object is present based on the global self-similarity descriptor. By using global self-similarity rather than local similarities more information is captured which can lead to better classification. In particular, it helps enable recognition of more distant self-similarities inherent in the object, and self-similarities present at any scale.

Abstract: Methods and apparatus are described for monocular 3D human pose estimation and tracking, which are able to recover poses of people in realistic street conditions captured using a monocular, potentially moving camera. Embodiments of the present invention provide a three-stage process involving estimating (10, 60, 110) a 3D pose of each of the multiple objects using an output of 2D tracking-by detection (50) and 2D viewpoint estimation (46). The present invention provides a sound Bayesian formulation to address the above problems. The present invention can provide articulated 3D tracking in realistic street conditions. The present invention provides methods and apparatus for people detection and 2D pose estimation combined with a dynamic motion prior. The present invention provides not only 2D pose estimation for people in side views, it goes beyond this by estimating poses in 3D from multiple viewpoints. The estimation of poses is done in monocular images, and does not require stereo images.

Abstract: The invention relates to a method and an object detection device for analyzing objects in the environment and/or scenes in the environment. The object detection device includes a data processing and/or evaluation device. In the data processing and/or evaluation device, image data (xt) is evaluated on the basis of a Conditional Random Field (CRF) model and the CRF model provides additional object nodes (otn) which take into account information from an object detector.

Abstract: Methods and apparatus are described for monocular 3D human pose estimation and tracking, which are able to recover poses of people in realistic street conditions captured using a monocular, potentially moving camera. Embodiments of the present invention provide a three-stage process involving estimating (10, 60, 110) a 3D pose of each of the multiple objects using an output of 2D tracking-by detection (50) and 2D viewpoint estimation (46). The present invention provides a sound Bayesian formulation to address the above problems. The present invention can provide articulated 3D tracking in realistic street conditions. The present invention provides methods and apparatus for people detection and 2D pose estimation combined with a dynamic motion prior. The present invention provides not only 2D pose estimation for people in side views, it goes beyond this by estimating poses in 3D from multiple viewpoints. The estimation of poses is done in monocular images, and does not require stereo images.

Abstract: An image processor (10) has a window selector for choosing a detection window within the image, and a self similarity computation part (40) for determining self-similarity information for a group of the pixels in any part of the detection window, to represent an amount of self-similarity of that group to other groups in any other part of the detector window, and for repeating the determination for groups in all parts of the detection window, to generate a global self similarity descriptor for the detection window. A classifier (50) is used for classifying whether an object is present based on the global self-similarity descriptor. By using global self-similarity rather than local similarities more information is captured which can lead to better classification. In particular, it helps enable recognition of more distant self-similarities inherent in the object, and self-similarities present at any scale.

Abstract: The invention relates to a method for automatically detecting a driving maneuver of a motor vehicle (A), in particular an overtaking maneuver or an evasive maneuver, in which the surroundings of the vehicle are covered and an electronic image thereof is created, the electronic image is used for the detection of a traffic lane and/or of a road as well as of-objects (B, C) in the surroundings of the vehicle, longitudinal-dynamics and lateral-dynamics movement information ({dot over (?)}, ?y, ?H, ?FL, ?FR, ?RL, ?RR) of motor vehicle (A) is determined, and the position ({circumflex over (X)}) of motor vehicle (A) is odometrically estimated on the basis of the data (bLane, yLane, ?, c0) of lane detection and/or road detection and/or of the movement information ({dot over (?)}, ?y, ?H, ?FL, ?FR, ?RL, ?RR) of motor vehicle (A), wherein the invention provides that a) the following indicator quantities are formed from the estimated position data ({circumflex over (X)}) of motor vehicle (A): a value of the latera

Abstract: Detection of an object of a specified object category in an image. With the method, it is provided that: (1) at least two detectors are provided which are respectively set up for the purpose of detecting an object of the specified object category with a specified object size, wherein object sizes differ for the detectors, (2) the image is evaluated by the detectors in order to check whether an object of the specified object category is located in the image, and (3) an object of the specified object category is detected in the image when on the basis of the evaluation of the image by at least one of the detectors it is determined that an object of the specified object category is located in the image. A system suitable for implementing the method for detecting an object of a specified object category in an image is also described.

Abstract: The invention relates to a method and an object detection device for analysing objects in the environment and/or scenes in the environment. The object detection device includes a data processing and/or evaluation device. In the data processing and/or evaluation device, image data (xt) is evaluated on the basis of a Conditional Random Field (CRF) model and the CRF model provides additional object nodes (otn) which take into account information from an object detector.