16,652 research outputs found
Finding the M-best consistent correspondences between 3D symmetric objects
No full textThis paper proposes a novel algorithm that resolves the underlying ambiguity in shape correspondences between symmetric objects. Due to the equivocal nature of symmetry, each part of a symmetric object may have two or more correspondence candidates in another symmetric object, which may induce local inconsistencies in the correspondence of parts or global ambiguities in shape matching. As an effective approach for resolving these symmetric ambiguities, we find multiple probable solutions for consistent shape correspondences between two 3D symmetric objects and let the user select one of them for an application-specific purpose. We formulate the problem of 3D symmetric object correspondences with a Markov Random Field (MRF) and iteratively search multiple solutions by excluding previously found solutions using Linear Programming (LP). The consistency of each solution is provided by four-point correspondences as high-order measurements in our MRF network, with each node corresponding to a point pair and each edge corresponding to a pair of point pairs. By leveraging the properties of the symmetry structure of the 3D object, we further reduce the complexity of our MRF network while efficiently handling high-order measurements. Finally, we evaluate the proposed algorithm using real-world symmetric object datasets. (c) 2012 Elsevier Ltd. All rights reserved.
Frequency comb-based time-domain tracking of AFM cantilever dynamics from picometre-scale noise to micron-scale nonlinear motion
No full textThe field of micro- and nano-mechanics has seen rapid advances driven by applications in sensing, microscopy, and precision instrumentation. Accurate, time-resolved characterization of mechanical dynamics is essential for understanding device behaviour and improving performance. However, conventional optical and electrical methods face trade-offs between sensitivity, linearity, and bandwidth, while frequency-domain approaches are limited in capturing transient dynamics. Here, we present a frequency comb-based time-domain tracking technique for directly observing the full-range dynamic motion of atomic force microscopy (AFM) micro-cantilevers. By leveraging electro-optic sampling between femtosecond optical pulses and ultra-precise photocurrent timing signals, our system enables real-time measurements across six orders of magnitude - from similar to 30 pm thermal fluctuations to similar to 20 mu m non-linear oscillations. The technique reveals complex behaviours including mode coupling, hysteresis, bifurcation, and transient modulation, while maintaining calibration fidelity through thermomechanical noise. This approach bridges the longstanding gap between ultra-sensitive and wide-range motion tracking, offering a powerful tool for studying nonlinear dynamics in micro- and nano-scale mechanical systems. Looking ahead, the method lays the groundwork for advances in high-resolution force sensing, AFM probe optimization, and the broader exploration of dynamic behaviour in precision microsystems.
Few-femtosecond jitter microwave signal generation from free-running mode-locked Er-fiber lasers
No full textNLNL: Negative Learning for Noisy Labels
No full textConvolutional Neural Networks (CNNs) provide excellent performance when used for image classification. The classical method of training CNNs is by labeling images in a supervised manner as in "input image belongs to this label" (Positive Learning; PL), which is a fast and accurate method if the labels are assigned correctly to all images. However, if inaccurate labels, or noisy labels, exist, training with PL will provide wrong information, thus severely degrading performance. To address this issue, we start with an indirect learning method called Negative Learning (NL), in which the CNNs are trained using a complementary label as in "input image does not belong to this complementary label." Because the chances of selecting a true label as a complementary label are low, NL decreases the risk of providing incorrect information. Furthermore, to improve convergence, we extend our method by adopting PL selectively, termed as Selective Negative Learning and Positive Learning (SelNLPL). PL is used selectively to train upon expected-to-be-clean data, whose choices become possible as NL progresses, thus resulting in superior performance of filtering out noisy data. With simple semi-supervised training technique, our method achieves state-of-the-art accuracy for noisy data classification, proving the superiority of SelNLPL's noisy data filtering ability
Worst Case Eye Estimation Method considering Power Noise on Tx Output Driver and Channel Crosstalk & ISI in Interposer-based 2.5D Interfaces
No full text
Optically Invisible Antenna‐on‐Display (AoD) Technologies: Review, Demonstration and Opportunities for Microwave, Millimeter‐Wave and Sub‐THz Wireless Applications
No full textThis paper provides a detailed overview of an optically invisible phased-array antenna-on-display (AoD) technology for microwave, millimeter-wave (mmWave), and sub-THz wireless scenarios such as wireless communication, radar, sensing etc. The fundamental AoD configuration and stack-up are introduced with numerical results, which proves the trade-off relationship between optical transmittance and RF conductivity. A transverse magnetic patch antenna for Wi-Fi and Bluetooth accessibility is implemented on the active display panel of a real-life smartwatch device. The fully functional smartwatch prototype including the electromagnetic wrist phantom features a total radiation efficiency of 22% at 2.4 GHz. The optically invisible phased-array AoD is fully integrated within the view area of high-resolution OLED display panels of mmWave 5G NR cellular handsets. The fabricated cellular handset prototype including optically invisible beamforming AoD features optical transmittance of more than 88 % and beam scanning of ± 30° with the estimated maximum EIRP of 14.4 dBm. The over-the-air test confirms that the fabricated phased-array AoD prototype satisfies the 3GPP's EVM performance requirement for 5G NR cellular devices. Future work collectively aims to serve as a catalyst for the emergence of future sub-THz 6G mobile devices.11Nscopu
Separability analysis and classification of rice fields using KOMPSAT-2 high resolution satellite imagery
No full text- …
