Cross-modality person re-identification is a challenging problem which retrieves a given pedestrian image in RGB modality among all the gallery images in infrared modality. The task can address the limitation of RGB-based person Re-ID in dark environments. Existing researches mainly focus on enlarging inter-class differences of feature to solve the problem. However, few studies investigate improving intra-class cross-modality similarity, which is important for this issue. In this paper, we propose a novel loss function, called Hetero-Center loss (HC loss) to reduce the intra-class cross-modality variations. Specifically, HC loss can supervise the network learning the cross-modality invariant information by constraining the intra-class center distance between two heterogenous modalities. With the joint supervision of Cross-Entropy (CE) loss and HC loss, the network is trained to achieve two vital objectives, inter-class discrepancy and intra-class cross-modality similarity as much as possible. Besides, we propose a simple and high-performance network architecture to learn local feature representations for cross-modality person re-identification, which can be a baseline for future research. Extensive experiments indicate the effectiveness of the proposed methods, which outperform state-of-the-art methods by a wide margin.
This paper studies the optimal least squares and minimax design and realization of digital synthesis filters for hybrid-filter-bank analog-to-digital converters (HFB ADCs) to meet a given spurious-free dynamic range (SFDR). The problem for designing finite-impulse-response synthesis filters is formulated as a second-order cone-programming problem, which is convex and allows linear and quadratic constraints such as peak aliasing error to be incorporated. The fixed coefficients of the designed synthesis filters are efficiently implemented using sum-of-power-of-two (SOPOT) coefficients, while the internal word length used for each intermediate data is minimized using geometric programming. The main sources of error are analyzed, and a new formula of SFDR in terms of these errors is derived. The effects of component variations of analog analysis filters on the HFB ADC are also addressed by means of two new robust HFB ADC design algorithms based on stochastic uncertainty and worst case uncertainty models. Design results show that the proposed approach offers more flexibility and better performance than conventional methods in achieving a given SFDR and that the robust design algorithms are more robust to parameter uncertainties than the nominal design in which the uncertainties are not taken into account.Index Terms-Analog-to-digital converters (ADCs), hybrid filter bank (HFB), multiplierless realization, parameter uncertainty, second-order cone programming (SOCP), synthesis filter design.
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.