IEEE Transactions on Pattern Analysis and Machine Intelligence

With the continual miniaturisation of sensors and processing nodes, Pedestrian Dead Reckoning (PDR) systems are becoming feasible options for indoor tracking. These use inertial and other sensors, often combined with domain-specific knowledge about walking, to track user movements. There is currently a wealth of relevant literature spread across different research communities. In this survey, a taxonomy of modern PDRs is developed and used to contextualise the contributions from different areas. Techniques for step detection, characterisation, inertial navigation and step-and-heading-based dead-reckoning are reviewed and compared. Techniques that incorporate building maps through particle filters are analysed, along with hybrid systems that use absolute position fixes to correct dead-reckoning output. In addition, consideration is given to the possibility of using smartphones as PDR sensing devices. The survey concludes that PDR techniques alone can offer good short- to medium- term tracking under certain circumstances, but that regular absolute position fixes from partner systems will be needed to ensure long-term operation and to cope with unexpected behaviours. It concludes by identifying a detailed list of challenges for PDR researchers.

A Survey of Indoor Inertial Positioning Systems for Pedestrians

Most machine-learning methods focus on classifying instances whose classes have already been seen in training. In practice, many applications require classifying instances whose classes have not been seen previously. Zero-shot learning is a powerful and promising learning paradigm, in which the classes covered by training instances and the classes we aim to classify are disjoint. In this paper, we provide a comprehensive survey of zero-shot learning. First of all, we provide an overview of zero-shot learning. According to the data utilized in model optimization, we classify zero-shot learning into three learning settings. Second, we describe different semantic spaces adopted in existing zero-shot learning works. Third, we categorize existing zero-shot learning methods and introduce representative methods under each category. Fourth, we discuss different applications of zero-shot learning. Finally, we highlight promising future research directions of zero-shot learning.

A Survey of Zero-Shot Learning: Settings, Methods, and Applications

Aim
Individuals with cerebral palsy (CP) have impaired movement due to a brain injury near birth. Understanding how neuromuscular control is altered in CP can provide insight into pathological movement. We sought to determine if individuals with CP demonstrate reduced complexity of neuromuscular control during gait compared with unimpaired individuals and if changes in control are related to functional ability.

Method
Muscle synergies during gait were retrospectively analyzed for 633 individuals (age range 3.9–70y): 549 with CP (hemiplegia, n=122; diplegia, n=266; triplegia, n=73; quadriplegia, n=88) and 84 unimpaired individuals. Synergies were calculated using non-negative matrix factorization from surface electromyography collected during previous clinical gait analyses. Synergy complexity during gait was compared with diagnosis subtype, functional ability, and clinical examination measures.

Result
Fewer synergies were required to describe muscle activity during gait in individuals with CP compared with unimpaired individuals. Changes in synergies were related to functional impairment and clinical examination measures including selective motor control, strength, and spasticity.

Interpretation
Individuals with CP use a simplified control strategy during gait compared with unimpaired individuals. These results were similar to synergies during walking among adult stroke survivors, suggesting similar neuromuscular control strategies between these clinical populations.

https://www.onlinelibrary.wiley.com/doi/pdf/10.1111/dmcn.12826

Muscle synergies and complexity of neuromuscular control during gait in cerebral palsy.

Device-free wireless localization and activity recognition (DFLAR) is a new technique, which could estimate the location and activity of a target by analyzing its shadowing effect on surrounding wireless links. This technique neither requires the target to be equipped with any device nor involves privacy concerns, which makes it an attractive and promising technique for many emerging smart applications. The key question of DFLAR is how to characterize the influence of the target on wireless signals. Existing work generally utilizes statistical features extracted from wireless signals, such as mean and variance in the time domain and energy as well as entropy in the frequency domain, to characterize the influence of the target. However, a feature suitable for distinguishing some activities or gestures may perform poorly when it is used to recognize other activities or gestures. Therefore, one has to manually design handcraft features for a specific application. Inspired by its excellent performance in extracting universal and discriminative features, in this paper, we propose a deep learning approach for realizing DFLAR. Specifically, we design a sparse autoencoder network to automatically learn discriminative features from the wireless signals and merge the learned features into a softmax-regression-based machine learning framework to realize location, activity, and gesture recognition simultaneously. Extensive experiments performed in a clutter indoor laboratory and an apartment with eight wireless nodes demonstrate that the DFLAR system using the learned features could achieve 0.85 or higher accuracy, which is better than the systems utilizing traditional handcraft features.

Device-Free Wireless Localization and Activity Recognition: A Deep Learning Approach

Zero-shot learning for visual recognition, e.g., object and action recognition, has recently attracted a lot of attention. However, it still remains challenging in bridging the semantic gap between visual features and their underlying semantics and transferring knowledge to semantic categories unseen during learning. Unlike most of the existing zero-shot visual recognition methods, we propose a stagewise bidirectional latent embedding framework of two subsequent learning stages for zero-shot visual recognition. In the bottom---up stage, a latent embedding space is first created by exploring the topological and labeling information underlying training data of known classes via a proper supervised subspace learning algorithm and the latent embedding of training data are used to form landmarks that guide embedding semantics underlying unseen classes into this learned latent space. In the top---down stage, semantic representations of unseen-class labels in a given label vocabulary are then embedded to the same latent space to preserve the semantic relatedness between all different classes via our proposed semi-supervised Sammon mapping with the guidance of landmarks. Thus, the resultant latent embedding space allows for predicting the label of a test instance with a simple nearest-neighbor rule. To evaluate the effectiveness of the proposed framework, we have conducted extensive experiments on four benchmark datasets in object and action recognition, i.e., AwA, CUB-200-2011, UCF101 and HMDB51. The experimental results under comparative studies demonstrate that our proposed approach yields the state-of-the-art performance under inductive and transductive settings.

/pdf/zero-shot-visual-recognition-via-bidirectional-latent-20nav5tvav.pdf

Zero-Shot Visual Recognition via Bidirectional Latent Embedding

Unsupervised domain adaptation aims to address the problem of classifying unlabeled samples from the target domain whilst labeled samples are only available from the source domain and the data distributions are different in these two domains. As a result, classifiers trained from labeled samples in the source domain suffer from significant performance drop when directly applied to the samples from the target domain. To address this issue, different approaches have been proposed to learn domain-invariant features or domain-specific classifiers. In either case, the lack of labeled samples in the target domain can be an issue which is usually overcome by pseudo-labeling. Inaccurate pseudo-labeling, however, could result in catastrophic error accumulation during learning. In this paper, we propose a novel selective pseudo-labeling strategy based on structured prediction. The idea of structured prediction is inspired by the fact that samples in the target domain are well clustered within the deep feature space so that unsupervised clustering analysis can be used to facilitate accurate pseudo-labeling. Experimental results on four datasets (i.e. Office-Caltech, Office31, ImageCLEF-DA and Office-Home) validate our approach outperforms contemporary state-of-the-art methods.

/pdf/unsupervised-domain-adaptation-via-structured-prediction-3rmptg440j.pdf

Unsupervised Domain Adaptation via Structured Prediction Based Selective Pseudo-Labeling

Zero-shot learning for visual recognition, e.g., object and action recognition, has recently attracted a lot of attention. However, it still remains challenging in bridging the semantic gap between visual features and their underlying semantics and transferring knowledge to semantic categories unseen during learning. Unlike most of the existing zero-shot visual recognition methods, we propose a stagewise bidirectional latent embedding framework to two subsequent learning stages for zero-shot visual recognition. In the bottom-up stage, a latent embedding space is first created by exploring the topological and labeling information underlying training data of known classes via a proper supervised subspace learning algorithm and the latent embedding of training data are used to form landmarks that guide embedding semantics underlying unseen classes into this learned latent space. In the top-down stage, semantic representations of unseen-class labels in a given label vocabulary are then embedded to the same latent space to preserve the semantic relatedness between all different classes via our proposed semi-supervised Sammon mapping with the guidance of landmarks. Thus, the resultant latent embedding space allows for predicting the label of a test instance with a simple nearest-neighbor rule. To evaluate the effectiveness of the proposed framework, we have conducted extensive experiments on four benchmark datasets in object and action recognition, i.e., AwA, CUB-200-2011, UCF101 and HMDB51. The experimental results under comparative studies demonstrate that our proposed approach yields the state-of-the-art performance under inductive and transductive settings.

Unsupervised domain adaptation aims to transfer knowledge from a source domain to a target domain so that the target domain data can be recognized without any explicit labelling information for this domain. One limitation of the problem setting is that testing data (despite no labels) from the target domain is needed during training, which prevents the trained model being directly applied to classify newly arrived test instances. We formulate a new cross-domain classification problem arising from real-world scenarios where labelled data are available for a subset of classes (known classes) in the target domain, and we expect to recognize new samples belonging to any class (known and unseen classes) once the model is learned. This is a generalized zero-shot learning problem where the side information comes from the source domain in the form of labelled samples instead of class-level semantic representations commonly used in traditional zero-shot learning. We present a unified domain adaptation framework for both unsupervised and zero-shot learning conditions. Our approach learns a joint subspace from source and target domains so that the projections of both data in the subspace can be domain invariant and easily separable. We use the supervised locality preserving projection (SLPP) as the enabling technique and conduct experiments under both unsupervised and zero-shot learning conditions, achieving state-of-the-art results on three domain adaptation benchmark datasets: Office-Caltech, Office31 and Office-Home.

Unifying Unsupervised Domain Adaptation and Zero-Shot Visual Recognition

A proper semantic representation for encoding side information is key to the success of zero-shot learning. In this paper, we explore two alternative semantic representations especially for zero-shot human action recognition: textual descriptions of human actions and deep features extracted from still images relevant to human actions. Such side information are accessible on Web with little cost, which paves a new way in gaining side information for large-scale zero-shot human action recognition. We investigate different encoding methods to generate semantic representations for human actions from such side information. Based on our zero-shot visual recognition method, we conducted experiments on UCF101 and HMDB51 to evaluate two proposed semantic representations. The results suggest that our proposed text- and image-based semantic representations outperform traditional attributes and word vectors considerably for zero-shot human action recognition. In particular, the image-based semantic representations yield the favourable performance even though the representation is extracted from a small number of images per class.

/pdf/alternative-semantic-representations-for-zero-shot-human-4y7dbbz7oo.pdf

Qian Wang

Papers

Zero-Shot Visual Recognition via Bidirectional Latent Embedding

Unsupervised Domain Adaptation via Structured Prediction Based Selective Pseudo-Labeling

Zero-Shot Visual Recognition via Bidirectional Latent Embedding

Unifying Unsupervised Domain Adaptation and Zero-Shot Visual Recognition

Alternative Semantic Representations for Zero-Shot Human Action Recognition