Humans are able to learn to recognize new objects even from a few examples. In contrast, training deep-learning-based object detectors requires huge amounts of annotated data. To avoid the need to acquire and annotate these huge amounts of data, few-shot object detection (FSOD) aims to learn from few object instances of new categories in the target domain. In this survey, we provide an overview of the state of the art in FSOD. We categorize approaches according to their training scheme and architectural layout. For each type of approach, we describe the general realization as well as concepts to improve the performance on novel categories. Whenever appropriate, we give short takeaways regarding these concepts in order to highlight the best ideas. Eventually, we introduce commonly used datasets and their evaluation protocols and analyze the reported benchmark results. As a result, we emphasize common challenges in evaluation and identify the most promising current trends in this emerging field of FSOD.

In order to meet the increasing demands of mobile service robot applications, a dedicated perception module is an essential requirement for the interaction with users in real-world scenarios. In particular, multi sensor fusion and human re-identification are recognized as active research fronts. Through this paper we contribute to the topic and present a modular detection and tracking system that models position and additional properties of persons in the surroundings of a mobile robot. The proposed system introduces a probability-based data association method that besides the position can incorporate face and color-based appearance features in order to realize a re-identification of persons when tracking gets interrupted. The system combines the results of various state-of-the-art image-based detection systems for person recognition, person identification and attribute estimation. This allows a stable estimate of a mobile robot’s user, even in complex, cluttered environments with long-lasting occlusions. In our benchmark, we introduce a new measure for tracking consistency and show the improvements when face and appearance-based re-identification are combined. The tracking system was applied in a real world application with a mobile rehabilitation assistant robot in a public hospital. The estimated states of persons are used for the user-centered navigation behaviors, e.g., guiding or approaching a person, but also for realizing a socially acceptable navigation in public environments.

Deep learning-based autoencoders represent a promising technology for use in network-based attack detection systems. They offer significant benefits in managing unknown network traces or novel attack signatures. Specifically, in the context of critical infrastructures, such as power supply systems, AI-based intrusion detection systems must meet stringent requirements concerning model accuracy and trustworthiness. For the intrusion response, the activation of suitable countermeasures can greatly benefit from additional transparency information (e.g., attack causes). Transformers represent the state of the art for learning from sequential data and provide important model insights through the widespread use of attention mechanisms. This paper introduces a two-stage transformer-based autoencoder for learning meaningful information from network traffic at the packet and sequence level. Based on this, we present a sequential attention weight perturbation method to explain benign and malicious network packets. We evaluate our method against benchmark models and expert-based explanations using the CIC-IDS-2017 benchmark dataset. The results show promising results in terms of detecting and explaining FTP and SSH brute-force attacks, highly outperforming the results of the benchmark model.

Fast vision-based object and person tracking is important for various applications in mobile robotics and Human-Robot Interaction. While current state-of-the-art methods use descriptive features for visual tracking, we propose a novel approach using a sparse template based feature set, which is drawn from homogeneous regions on the object to be tracked. Using only a small number of simple features, without complex descriptors in combination with logarithmic-search, the tracker performs at hyper-real-time on HD-images without the use of parallelized hardware. Detailed benchmark experiments show that it outperforms most other state-of-the-art approaches for real-time object and person tracking in quality and runtime. In the experiments we also show the robustness of the tracker and evaluate the effects of different initialization methods, feature sets, and parameters on the tracker. Although we focus on the scenario of person and object tracking in robot applications, the proposed tracker can be used for a variety of other tracking tasks.

Object detection is a crucial capability of autonomous agents for human–robot collaboration, as it facilitates the identification of the current processing state. In industrial scenarios, it is uncommon to have comprehensive knowledge of all the objects involved in a given task. Furthermore, training during deployment is not a viable option. Consequently, there is a need for a detector that is able to adapt to novel objects during deployment without the necessity of retraining or fine-tuning on novel data. To achieve this, we propose to exploit the ability of discriminative embeddings learned by an object re-identification model to generalize to unknown categories described by a few shots. To do so, we extract object crops with a class-agnostic detector and then compare the object features with the prototypes of the novel objects. Moreover, we demonstrate that the embedding is also effective for predicting regions of interest, which narrows the search space of the class-agnostic detector and, consequently, increases processing speed. The effectiveness of our approach is evaluated in an assembly scenario, wherein the majority of objects belong to categories distinct from those present in the training datasets. Our experiments demonstrate that, in this scenario, our approach outperforms the current best few-shot object-detection approach DE-ViT, which also does not perform fine-tuning on novel data, in terms of both detection capability and inference speed.