Search

Crowd counting has received increasing attention in the field of video surveillance and urban security system. However, many previous models are prone to poor generalization capability to unknown samples when limited labeled samples are available. To improve or mitigate the above weakness, we develop a novel Pseudo-label Growth Dictionary Pair Learning (PG-DPL) method for crowd counting. To be exact, we treat crowd counting as a task of classification and leverage dictionary learning-based (DL) strategy to target the task. Considering that being short of diverse training samples and imbalanced distribution across different classes in crowd scene inevitably result in large prediction deviation caused by the DL model, we propose to apply pseudo-label growth (PG) and adaptive dictionary size (ADS) to improve the accuracy of crowd counting with limited labeled samples. In the proposed method, PG optimizes the initial prediction via reconstructing the discriminant term to improve the robustness of learned dictionary, while ADS explores the imbalanced distribution among different classes to adapt to the size of class-specific dictionary. Extensive validation experiments on five benchmark databases indicate that the proposed PG-DPL can achieve compelling performance compared to other state-of-the-art methods.

In this Letter, the authors present a novel graph clustering-based method for crowd counting only using very limited labelled samples. Based on an intuitional observation that the distribution of low-level features of a specific scene containing the same or similar number of pedestrians are close to each other in the feature space, the authors adopt a first neighbour propagation (FNP) based clustering method to divide all unlabelled data into different groups. Next, an active sampling learning strategy that measures representativeness and diversity of the training data is used to obtain a few informative samples for annotation. Finally, the counts of those labelled informative samples are effectively propagated to predict the unlabelled samples in the constructed clusters by FNP-based clustering. The compelling results on two benchmark datasets demonstrate that the proposed method is not only effective to estimate crowd counts with very few labelled samples but also applicable to annotate a large number of unknown video frames for scene-specific crowd counting models.

In a multiple-input multiple-output (MIMO) radar, it is desired to maximize the power radiation in an angular region of interest, which, however, is hardly achievable because a set of orthogonal waveforms is employed at the engineering. Different from traditional colocated MIMO radar, space-time coding array (STCA) transmits an identical waveform with a tiny time delay circulating across array elements. It can provide a simple way to achieve the full angular coverage with a stable gain. In this paper, a new scheme referred to as chirp-STCA is proposed to synthesize the desired beampattern flexibly. With the quantitative relationship between the time delay and the span of beam spreading, the transmit beampattern synthesis issue is converted to the design of time delay value which can be readily realized based on the framework of chirp-STCA. As a result, the desired angular sector coverage of the radiation power can be realized by setting a proper time delay. The proposed chirp-STCA can provide many desired features including low range sidelobe level, low computational complexity, and constant modulus. In addition, the range-angle properties are examined by analyzing the transmit beampattern response and the ambiguity function. Numerical results are presented to demonstrate the effectiveness and simplicity of the chirp-STCA scheme for the transmit beampattern synthesis.

Crowd counting is one of the most paramount tasks for safety and security. Many existing methods mainly focus on the predicted accuracy but ignore the efficiency, which hinders their applications in practice. Moreover, their performance heavily depends on the learning from a large number of labeled scene data, which is cost-expensive for crowd counting. In this paper, we present a novel crowd counting approach called semi-supervised manifold embedding (SSME) to address the above weaknesses. In the newly proposed method, we formulate the crowd counting as a semi-supervised classification problem and learn a linear mapping from the high-dimensional scene feature space to the low-dimension label space by simultaneously imposing the label fitness and the manifold smoothness, where the learned linear mapping facilitates the efficiency of crowd counting. In order to alleviate the issue that most supervised approaches to crowd counting require sufficient labeled data for improving the performance, we exploit the first neighbor propagation to select informative samples in the proposed SSME-based algorithm. Thorough validation experiments on three challenging benchmark datasets indicate that the proposed method is capable of achieving more impressive prediction accuracy on the number of pedestrians in a monitoring scene than other state-of-the-art competitors.