Multiple instance learning (MIL) is a form of weakly supervised learning where training instances are arranged in sets, called bags, and a label is provided for the entire bag. This formulation is gaining interest because it naturally fits various problems and allows to leverage weakly labeled data. Consequently, it has been used in diverse application fields such as computer vision and document classification. However, learning from bags raises important challenges that are unique to MIL. This paper provides a comprehensive survey of the characteristics which define and differentiate the types of MIL problems. Until now, these problem characteristics have not been formally identified and described. As a result, the variations in performance of MIL algorithms from one data set to another are difficult to explain. In this paper, MIL problem characteristics are grouped into four broad categories: the composition of the bags, the types of data distribution, the ambiguity of instance labels, and the task to be performed. Methods specialized to address each category are reviewed. Then, the extent to which these characteristics manifest themselves in key MIL application areas are described. Finally, experiments are conducted to compare the performance of 16 state-of-the-art MIL methods on selected problem characteristics. This paper provides insight on how the problem characteristics affect MIL algorithms, recommendations for future benchmarking and promising avenues for research.
A growing number of applications, e.g. video surveillance and medical image analysis, require training recognition systems from large amounts of weakly annotated data while some targeted interactions with a domain expert are allowed to improve the training process. In such cases, active learning (AL) can reduce labeling costs for training a classifier by querying the expert to provide the labels of most informative instances. This paper focuses on AL methods for instance classification problems in multiple instance learning (MIL), where data is arranged into sets, called bags, that are weakly labeled. Most AL methods focus on single instance learning problems. These methods are not suitable for MIL problems because they cannot account for the bag structure of data. In this paper, new methods for bag-level aggregation of instance informativeness are proposed for multiple instance active learning (MIAL). The aggregated informativeness method identifies the most informative instances based on classifier uncertainty, and queries bags incorporating the most information. The other proposed method, called clusterbased aggregative sampling, clusters data hierarchically in the instance space. The informativeness of instances is assessed by considering bag labels, inferred instance labels, and the proportion of labels that remain to be discovered in clusters. Both proposed methods significantly outperform reference methods in extensive experiments using benchmark data from several application domains. Results indicate that using an appropriate strategy to address MIAL problems yields a significant reduction in the number of queries needed to achieve the same level of performance as single instance AL methods.Recent years have witnessed substantial advances of machine learning techniques that promise to address many complex large-scale problems that were previously thought intractable. However, in many applications, annotating enough representative training data to train a recognition system is costly, and in such cases, one can resort to AL to reduce the annotation burden [1,2]. Moreover, several applications allow to leverage some targeted interactions with human experts, as needed, to label informative data and drive the training process. AL has been used in various applications to reduce the cost of annotations, e.g., in medical image segmentation [3], text classification [4,5] and visual object detection [6].Alternatively, the cost of annotations can be reduced through weakly supervised learning. It generalizes many kinds of learning paradigms including semi-supervised learning and MIL in partially observable environments or learning from uncertain labels. With MIL, training instances are grouped in sets (commonly referred to as bags), and a label is only provided for an entire set, but not for each individual instance. MIL has also been shown to efficiently reduce annotation costs in several
Abstract-This paper presents a two-stage hierarchical method for play-break detection on non-edited team sports video feed. Unlike most existing methods, this algorithm uses modern action and event recognition method thus does not rely on production cues of broadcast feeds, but instead concentrates on the content of the video. Moreover, the method does not require player tracking, can be used in real-time and can be easily adapted to different sports. In the first stage, bag-of-words event detectors are trained to recognize key events such as line changes, face-offs and preliminary play-breaks. In the second stage, the output of the detectors along with a novel feature based on the number of detected spatio-temporal interest points are used to create a context descriptor. The final classification is performed on this context descriptor. Experiments demonstrate the benefits of using this context descriptor by reducing the frame classification error by 18% when compared to the baseline method. The efficiency of the proposed method is demonstrated on a real hockey game (accuracy over 88%).
A novel Air-to-Ground (ATG) communication system, based on adaptive modulation and beamforming enabled by Automatic Dependent Surveillance-Broadcast (ADS-B) and multilateration technique is presented in this work. From an aircraft geolocation perspective, the proposed multilateration technique uses the Time-Difference-of-Arrival (TDOA), Angleof-Arrival (AOA), and Frequency-Difference-of-Arrival (FDOA) features within the ADS-B signal to implement the hybrid geolocation mechanism. Moreover, this hybrid mechanism aims at the optimal selection of multilateration sensors to provide a precise aircraft geolocation estimate by minimizing the Geometric Dilution of Precision (GDOP) metric and also imparts significant resilience to the current ADS-B based geolocation framework to withstand any form of attack involving aircraft-impersonation and ADS-B message infringement. From an ATG communication perspective, the ground Base Stations (BSs) can use this hybrid aircraft geolocation estimate to dynamically adapt their modulation parameters and transmission beampattern, in an effortto provide a high data rate secure ATG communication link. Additionally, we develop a hardware prototype which is highly accurate in estimating the AOA data and facilitating TDOA, FDOA extraction associated with the received ADS-B signal. This hardware setup for the ADS-B based ATG system is analytically established and validated with commercially available universalsoftware-defined-radio-peripheral (USRP) units. This hardware setup displays a 1.5 • AOA estimation accuracy, whereas the simulated geolocation accuracy is approximately 30 m over 100 nautical miles for a typical aircraft trajectory. The adaptive modulation and beamforming approach assisted by the proposed GDOP minimization based multilateration strategy achieves significant enhancement in throughput and reduction in packet error rate.
Abstract-A cognitive detect and avoid radar system based on chaotic UWB-MIMO waveform design to enable autonomous UAV navigation is presented. A Dirichlet-Process-Mixture-Model (DPMM) based Bayesian clustering approach to discriminate extended targets and a Change-Point (CP) detection algorithm are applied for the autonomous tracking and identification of potential collision threats. A DPMM based clustering mechanism does not rely upon any a priori target scene assumptions and facilitates online multivariate data clustering/classification for an arbitrary number of targets. Furthermore, this radar system utilizes a cognitive mechanism to select efficient c haotic waveforms to facilitate enhanced target detection and discrimination. We formulate the CP mechanism for the online tracking of target trajectories which present a collision threat to the UAV navigation and thus we supplement the conventional Kalman filter based tracking. Simulation results demonstrate a significant performance improvement for the DPMM-CP assisted detection as compared with direct generalized likelihood ratio based detection. Specifically, w e o bserve a 4 d B p erformance g ain i n target detection over conventional fixed U WB w aveforms a nd superior collision avoidance capability offered by the joint DPMM-CP mechanism.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.