arXiv subjects

Xudong Jiang

Publications and source records attributed to Xudong Jiang.

113 records · Page 7Linked to original sources

arXiv2019

Feature Boosting Network For 3D Pose Estimation

In this paper, a feature boosting network is proposed for estimating 3D hand pose and 3D body pose from a single RGB image. In this method, the features learned by the convolutional layers are boosted with a new long short-term dependence-aware (LSTD) module, which enables the intermediate convolutional feature maps to perceive the graphical long short-term dependency among different hand (or body) parts using the designed Graphical ConvLSTM. Learning a set of features that are reliable and discriminatively representative of the pose of a hand (or body) part is difficult due to the ambiguities, texture and illumination variation, and self-occlusion in the real application of 3D pose estimation. To improve the reliability of the features for representing each body part and enhance the LSTD module, we further introduce a context consistency gate (CCG) in this paper, with which the convolutional feature maps are modulated according to their consistency with the context representations. We evaluate the proposed method on challenging benchmark datasets for 3D hand pose estimation and 3D full body pose estimation. Experimental results show the effectiveness of our method that achieves state-of-the-art performance on both of the tasks.

cs.CV↗

arXiv2018

DCI: Discriminative and Contrast Invertible Descriptor

Local feature descriptors have been widely used in fine-grained visual object search thanks to their robustness in scale and rotation variation and cluttered background. However, the performance of such descriptors drops under severe illumination changes. In this paper, we proposed a Discriminative and Contrast Invertible (DCI) local feature descriptor. In order to increase the discriminative ability of the descriptor under illumination changes, a Laplace gradient based histogram is proposed. A robust contrast flipping estimate is proposed based on the divergence of a local region. Experiments on fine-grained object recognition and retrieval applications demonstrate the superior performance of DCI descriptor to others.

cs.CV↗

arXiv2018

Interest Point Detection based on Adaptive Ternary Coding

In this paper, an adaptive pixel ternary coding mechanism is proposed and a contrast invariant and noise resistant interest point detector is developed on the basis of this mechanism. Every pixel in a local region is adaptively encoded into one of the three statuses: bright, uncertain and dark. The blob significance of the local region is measured by the spatial distribution of the bright and dark pixels. Interest points are extracted from this blob significance measurement. By labeling the statuses of ternary bright, uncertain, and dark, the proposed detector shows more robustness to image noise and quantization errors. Moreover, the adaptive strategy for the ternary cording, which relies on two thresholds that automatically converge to the median of the local region in measurement, enables this coding to be insensitive to the image local contrast. As a result, the proposed detector is invariant to illumination changes. The state-of-the-art results are achieved on the standard datasets, and also in the face recognition application.

cs.CV↗

arXiv2012

An Ultra-Low Noise Telecom Wavelength Free Running Single Photon Detector Using Negative Feedback Avalanche Diode

It is challenging to implement genuine free running single photon detectors for the 1550 nm wavelength range with simultaneously high detection efficiency (DE), low dark noise, and good time resolution. We report a novel read out system for the signals from a negative feedback avalanche diode (NFAD) which allows useful operation of these devices at a temperature of 193 K and results in very low dark counts (~100 CPS), good time jitter (~30 ps), and good DE (~10%). We characterized two NFADs with a time correlation method using photons generated from weak coherent pulses (WCP) and photon pairs produced by spontaneous parametric down conversion (SPDC). The inferred detector efficiencies for both types of photon sources agree with each other. The best noise equivalent power of the device is estimated to be 8.1 x 10^(-18) W Hz^(-1/2), more than 10 times better than typical InP/InGaAs SPADs show in free running mode. The afterpulsing probability was found to be less than 0.1% per ns at the optimized operating point. In addition, we studied the performance of an entanglement-based quantum key distribution (QKD) using these detectors and develop a model for the quantum bit error rate (QBER) that incorporates the afterpulsing coefficients. We verified experimentally that using these NFADs it is feasible to implement QKD over 400 km of telecom fibre. Our NFAD photon detector system is very simple, and is well suited for single-photon applications where ultra-low noise and free-running operation is required, and some afterpulsing can be tolerated.

physics.ins-det↗

arXiv2012

Free Running Single Photon Detection based on a negative feedback InGaAs APD

InGaAs/InP-based semiconductor avalanche photodiode are usually employed for single-photon counting at telecom wavelength. However they are affected by afterpulsing which limits the diode performance. Recently, Princeton Lightwave has commercialised a diode integrating monolithically a feedback resistor. This solution effectively quenches the avalanche and drastically reduces afterpulsing. Here, we report the development and characterization of a detector module based on this diode, implementing an active hold-off circuit which further reduces the afterpulsing and notably improves the detector performances. We demonstrate free-running operation with 600 Hz dark count rate at 10% detection efficiency. We also improved the standard double-window technique for the afterpulsing characterization. Our algorithm implemented by a FPGA allows to put the APD in a well-defined initial condition and to measure the impact of the higher order afterpulses.

physics.ins-det↗