A unified neural network structure is presented for joint 3D object detection and point cloud segmentation in this paper. We leverage rich supervision from both detection and segmentation labels rather than using just one of them. In addition, an ext
ension based on single-stage object detectors is proposed based on the implicit function widely used in 3D scene and object understanding. The extension branch takes the final feature map from the object detection module as input, and produces an implicit function that generates semantic distribution for each point for its corresponding voxel center. We demonstrated the performance of our structure on nuScenes-lidarseg, a large-scale outdoor dataset. Our solution achieves competitive results against state-of-the-art methods in both 3D object detection and point cloud segmentation with little additional computation load compared with object detection solutions. The capability of efficient weakly supervision semantic segmentation of the proposed method is also validated by experiments.
In this paper, we propose an orthogonal frequency division multiplexing (OFDM)-based generalized optical quadrature spatial modulation (GOQSM) technique for multiple-input multiple-output optical wireless communication (MIMO-OWC) systems. Considering
the error propagation and noise amplification effects when applying maximum likelihood and maximum ratio combining (ML-MRC)-based detection, we further propose a deep neural network (DNN)-aided detection for OFDM-based GOQSM systems. The proposed DNN-aided detection scheme performs the GOQSM detection in a joint manner, which can efficiently eliminate the adverse effects of both error propagation and noise amplification. The obtained simulation results successfully verify the superiority of the deep learning-aided OFDM-based GOQSM technique for high-speed MIMO-OWC systems.
This paper proposes a method to extract the position and pose of vehicles in the 3D world from a single traffic camera. Most previous monocular 3D vehicle detection algorithms focused on cameras on vehicles from the perspective of a driver, and assum
ed known intrinsic and extrinsic calibration. On the contrary, this paper focuses on the same task using uncalibrated monocular traffic cameras. We observe that the homography between the road plane and the image plane is essential to 3D vehicle detection and the data synthesis for this task, and the homography can be estimated without the camera intrinsics and extrinsics. We conduct 3D vehicle detection by estimating the rotated bounding boxes (r-boxes) in the birds eye view (BEV) images generated from inverse perspective mapping. We propose a new regression target called textit{tailed~r-box} and a textit{dual-view} network architecture which boosts the detection accuracy on warped BEV images. Experiments show that the proposed method can generalize to new camera and environment setups despite not seeing imaged from them during training.
We study the localization properties of generalized, two- and three-dimensional Lieb lattices, $mathcal{L}_2(n)$ and $mathcal{L}_3(n)$, $n= 1, 2, 3$ and $4$, at energies corresponding to flat and dispersive bands using the transfer matrix method (TMM
) and finite size scaling (FSS). We find that the scaling properties of the flat bands are different from scaling in dispersive bands for all $mathcal{L}_d(n)$. For the $d=3$ dimensional case, states are extended for disorders $W$ down to $W=0.01 t$ at the flat bands, indicating that the disorder can lift the degeneracy of the flat bands quickly. The phase diagram with periodic boundary condition for $mathcal{L}_3(1)$ looks similar to the one for hard boundaries. We present the critical disorder $W_c$ at energy $E=0$ and find a decreasing $W_c$ for increasing $n$ for $mathcal{L}_3(n)$, up to $n=3$. Last, we show a table of FSS parameters including so-called irrelevant variables; but the results indicate that the accuracy is too low to determine these reliably. end{abstract}
The existed theories and methods for calculating interfacial thermal conductance of solid-solid interface lead to diverse values that deviate from experimental measurements. In this letter, We propose a model to estimate the ITC at high temperature w
ithout comprehensive calculations, where the interface between two dissimilar solids can be treated as an amorphous thin layer and the coordination number density across interface becomes a key parameter. Our model predicts that the ITCs of various interfaces at 300K are in a narrow range: 10$^{7}$W m$^{-2}$K$^{-1}$ $sim $10$^{9}$ W m$^{-2}$ K$^{-1}$, which is in good agreement with the experimental measurement.
Automated infection measurement and COVID-19 diagnosis based on Chest X-ray (CXR) imaging is important for faster examination. We propose a novel approach, called DRR4Covid, to learn automated COVID-19 diagnosis and infection segmentation on CXRs fro
m digitally reconstructed radiographs (DRRs). DRR4Covid comprises of an infection-aware DRR generator, a classification and/or segmentation network, and a domain adaptation module. The infection-aware DRR generator is able to produce DRRs with adjustable strength of radiological signs of COVID-19 infection, and generate pixel-level infection annotations that match the DRRs precisely. The domain adaptation module is introduced to reduce the domain discrepancy between DRRs and CXRs by training networks on unlabeled real CXRs and labeled DRRs together.We provide a simple but effective implementation of DRR4Covid by using a domain adaptation module based on Maximum Mean Discrepancy (MMD), and a FCN-based network with a classification header and a segmentation header. Extensive experiment results have confirmed the efficacy of our method; specifically, quantifying the performance by accuracy, AUC and F1-score, our network without using any annotations from CXRs has achieved a classification score of (0.954, 0.989, 0.953) and a segmentation score of (0.957, 0.981, 0.956) on a test set with 794 normal cases and 794 positive cases. Besides, we estimate the sensitive of X-ray images in detecting COVID-19 infection by adjusting the strength of radiological signs of COVID-19 infection in synthetic DRRs. The estimated detection limit of the proportion of infected voxels in the lungs is 19.43%, and the estimated lower bound of the contribution rate of infected voxels is 20.0% for significant radiological signs of COVID-19 infection. Our codes will be made publicly available at https://github.com/PengyiZhang/DRR4Covid.
Digital image watermarking is the process of embedding and extracting a watermark covertly on a cover-image. To dynamically adapt image watermarking algorithms, deep learning-based image watermarking schemes have attracted increased attention during
recent years. However, existing deep learning-based watermarking methods neither fully apply the fitting ability to learn and automate the embedding and extracting algorithms, nor achieve the properties of robustness and blindness simultaneously. In this paper, a robust and blind image watermarking scheme based on deep learning neural networks is proposed. To minimize the requirement of domain knowledge, the fitting ability of deep neural networks is exploited to learn and generalize an automated image watermarking algorithm. A deep learning architecture is specially designed for image watermarking tasks, which will be trained in an unsupervised manner to avoid human intervention and annotation. To facilitate flexible applications, the robustness of the proposed scheme is achieved without requiring any prior knowledge or adversarial examples of possible attacks. A challenging case of watermark extraction from phone camera-captured images demonstrates the robustness and practicality of the proposal. The experiments, evaluation, and application cases confirm the superiority of the proposed scheme.
Radiological image is currently adopted as the visual evidence for COVID-19 diagnosis in clinical. Using deep models to realize automated infection measurement and COVID-19 diagnosis is important for faster examination based on radiological imaging.
Unfortunately, collecting large training data systematically in the early stage is difficult. To address this problem, we explore the feasibility of learning deep models for COVID-19 diagnosis from a single radiological image by resorting to synthesizing diverse radiological images. Specifically, we propose a novel conditional generative model, called CoSinGAN, which can be learned from a single radiological image with a given condition, i.e., the annotations of the lung and COVID-19 infection. Our CoSinGAN is able to capture the conditional distribution of visual finds of COVID-19 infection, and further synthesize diverse and high-resolution radiological images that match the input conditions precisely. Both deep classification and segmentation networks trained on synthesized samples from CoSinGAN achieve notable detection accuracy of COVID-19 infection. Such results are significantly better than the counterparts trained on the same extremely small number of real samples (1 or 2 real samples) by using strong data augmentation, and approximate to the counterparts trained on large dataset (2846 real images). It confirms our method can significantly reduce the performance gap between deep models trained on extremely small dataset and on large dataset, and thus has the potential to realize learning COVID-19 diagnosis from few radiological images in the early stage of COVID-19 pandemic. Our codes are made publicly available at https://github.com/PengyiZhang/CoSinGAN.
We propose adversarial constrained-CNN loss, a new paradigm of constrained-CNN loss methods, for weakly supervised medical image segmentation. In the new paradigm, prior knowledge is encoded and depicted by reference masks, and is further employed to
impose constraints on segmentation outputs through adversarial learning with reference masks. Unlike pseudo label methods for weakly supervised segmentation, such reference masks are used to train a discriminator rather than a segmentation network, and thus are not required to be paired with specific images. Our new paradigm not only greatly facilitates imposing prior knowledge on networks outputs, but also provides stronger and higher-order constraints, i.e., distribution approximation, through adversarial learning. Extensive experiments involving different medical modalities, different anatomical structures, different topologies of the object of interest, different levels of prior knowledge and weakly supervised annotations with different annotation ratios is conducted to evaluate our ACCL method. Consistently superior segmentation results over the size constrained-CNN loss method have been achieved, some of which are close to the results of full supervision, thus fully verifying the effectiveness and generalization of our method. Specifically, we report an average Dice score of 75.4% with an average annotation ratio of 0.65%, surpassing the prior art, i.e., the size constrained-CNN loss method, by a large margin of 11.4%. Our codes are made publicly available at https://github.com/PengyiZhang/ACCL.
The microscopic mechanism of thermal transport in liquids and amorphous solids has been an outstanding problem for a long time. There have been several different approaches to explain the thermal conductivities for these systems, for example, the Bri
dgmans formula for simple liquids, the concept of the minimum thermal conductivity for amorphous solids, and the thermal resistance network model for amorphous polymers. Here, we present a ubiquitous formula to explain the thermal conductivities of liquids and amorphous solids in a unified way. The calculated thermal conductivities using this formula without fitting parameters are in excellent agreement with the experimental data for these systems. Our formula is not only providing detailed implications on microscopic mechanisms of heat transfer in these systems, but also solves the discrepancies between existing formulae and experimental data.
