arXiv subjects

Peng Yin

Publications and source records attributed to Peng Yin.

At least 55 records · Page 3Linked to original sources

arXiv2021

Advancing Self-supervised Monocular Depth Learning with Sparse LiDAR

Self-supervised monocular depth prediction provides a cost-effective solution to obtain the 3D location of each pixel. However, the existing approaches usually lead to unsatisfactory accuracy, which is critical for autonomous robots. In this paper, we propose FusionDepth, a novel two-stage network to advance the self-supervised monocular dense depth learning by leveraging low-cost sparse (e.g. 4-beam) LiDAR. Unlike the existing methods that use sparse LiDAR mainly in a manner of time-consuming iterative post-processing, our model fuses monocular image features and sparse LiDAR features to predict initial depth maps. Then, an efficient feed-forward refine network is further designed to correct the errors in these initial depth maps in pseudo-3D space with real-time performance. Extensive experiments show that our proposed model significantly outperforms all the state-of-the-art self-supervised methods, as well as the sparse-LiDAR-based methods on both self-supervised monocular depth prediction and completion tasks. With the accurate dense depth prediction, our model outperforms the state-of-the-art sparse-LiDAR-based method (Pseudo-LiDAR++) by more than 68% for the downstream task monocular 3D object detection on the KITTI Leaderboard. Code is available at https://github.com/AutoAILab/FusionDepth

cs.CV↗

arXiv2021

AdaFusion: Visual-LiDAR Fusion with Adaptive Weights for Place Recognition

Recent years have witnessed the increasing application of place recognition in various environments, such as city roads, large buildings, and a mix of indoor and outdoor places. This task, however, still remains challenging due to the limitations of different sensors and the changing appearance of environments. Current works only consider the use of individual sensors, or simply combine different sensors, ignoring the fact that the importance of different sensors varies as the environment changes. In this paper, an adaptive weighting visual-LiDAR fusion method, named AdaFusion, is proposed to learn the weights for both images and point cloud features. Features of these two modalities are thus contributed differently according to the current environmental situation. The learning of weights is achieved by the attention branch of the network, which is then fused with the multi-modality feature extraction branch. Furthermore, to better utilize the potential relationship between images and point clouds, we design a twostage fusion approach to combine the 2D and 3D attention. Our work is tested on two public datasets, and experiments show that the adaptive weights help improve recognition accuracy and system robustness to varying environments.

cs.CV↗

arXiv2021

Certification of Genuine Multipartite Entanglement with General and Robust Device-independent Witnesses

Genuine multipartite entanglement represents the strongest type of entanglement, which is an essential resource for quantum information processing. Standard methods to detect genuine multipartite entanglement, e.g., entanglement witnesses, state tomography, or quantum state verification, require full knowledge of the Hilbert space dimension and precise calibration of measurement devices, which are usually difficult to acquire in an experiment. The most radical way to overcome these problems is to detect entanglement solely based on the Bell-like correlations of measurement outcomes collected in the experiment, namely, device-independently (DI). However, it is difficult to certify genuine entanglement of practical multipartite states in this way, and even more difficult to quantify it, due to the difficulty to identify optimal multipartite Bell inequalities and protocols tolerant to state impurity. In this work, we explore a general and robust DI method which can be applied to various realistic multipartite quantum state in arbitrary finite dimension, while merely relying on bipartite Bell inequalities. Our method allows us both to certify the presence of genuine multipartite entanglement and to quantify it. Several important classes of entangled states are tested with this method, leading to the detection of genuinely entangled states. We also certify genuine multipartite entanglement in weakly-entangled GHZ states, thus showing that the method applies equally well to less standard states.

quant-ph↗

arXiv2021

PSE-Match: A Viewpoint-free Place Recognition Method with Parallel Semantic Embedding

Accurate localization on autonomous driving cars is essential for autonomy and driving safety, especially for complex urban streets and search-and-rescue subterranean environments where high-accurate GPS is not available. However current odometry estimation may introduce the drifting problems in long-term navigation without robust global localization. The main challenges involve scene divergence under the interference of dynamic environments and effective perception of observation and object layout variance from different viewpoints. To tackle these challenges, we present PSE-Match, a viewpoint-free place recognition method based on parallel semantic analysis of isolated semantic attributes from 3D point-cloud models. Compared with the original point cloud, the observed variance of semantic attributes is smaller. PSE-Match incorporates a divergence place learning network to capture different semantic attributes parallelly through the spherical harmonics domain. Using both existing benchmark datasets and two in-field collected datasets, our experiments show that the proposed method achieves above 70% average recall with top one retrieval and above 95% average recall with top ten retrieval cases. And PSE-Match has also demonstrated an obvious generalization ability with a limited training dataset.

cs.CV↗

arXiv2021

Exact solution of the Brueckner-Bethe-Goldstone equation with three-body forces in nuclear matter

An exact treatment of the operators Q/e(ω) and the total momentum is adopted to solve the nuclear matter Bruecker-Bethe-Goldstone equation with two- and three-body forces. The single-particle potential, equation of state and nucleon effective mass are calculated from the exact G-matrix. The results are compared with those obtained under the angle-average approximation and the angle-average approximation with total momentum approximation. It is found that the angle-average procedure, whereas preventing huge calculations of coupled channels, nevertheless provides a fairly accurate approximation. On the contrary, the total momentum approximation turns out to be quite inaccurate compared to its exact counterpart.

nucl-th↗

arXiv2021

i3dLoc: Image-to-range Cross-domain Localization Robust to Inconsistent Environmental Conditions

We present a method for localizing a single camera with respect to a point cloud map in indoor and outdoor scenes. The problem is challenging because correspondences of local invariant features are inconsistent across the domains between image and 3D. The problem is even more challenging as the method must handle various environmental conditions such as illumination, weather, and seasonal changes. Our method can match equirectangular images to the 3D range projections by extracting cross-domain symmetric place descriptors. Our key insight is to retain condition-invariant 3D geometry features from limited data samples while eliminating the condition-related features by a designed Generative Adversarial Network. Based on such features, we further design a spherical convolution network to learn viewpoint-invariant symmetric place descriptors. We evaluate our method on extensive self-collected datasets, which involve \textit{Long-term} (variant appearance conditions), \textit{Large-scale} (up to $2km$ structure/unstructured environment), and \textit{Multistory} (four-floor confined space). Our method surpasses other current state-of-the-arts by achieving around $3$ times higher place retrievals to inconsistent environments, and above $3$ times accuracy on online localization. To highlight our method's generalization capabilities, we also evaluate the recognition across different datasets. With a single trained model, i3dLoc can demonstrate reliable visual localization in random conditions.

cs.CV↗

arXiv2021

3D Segmentation Learning from Sparse Annotations and Hierarchical Descriptors

One of the main obstacles to 3D semantic segmentation is the significant amount of endeavor required to generate expensive point-wise annotations for fully supervised training. To alleviate manual efforts, we propose GIDSeg, a novel approach that can simultaneously learn segmentation from sparse annotations via reasoning global-regional structures and individual-vicinal properties. GIDSeg depicts global- and individual- relation via a dynamic edge convolution network coupled with a kernelized identity descriptor. The ensemble effects are obtained by endowing a fine-grained receptive field to a low-resolution voxelized map. In our GIDSeg, an adversarial learning module is also designed to further enhance the conditional constraint of identity descriptors within the joint feature distribution. Despite the apparent simplicity, our proposed approach achieves superior performance over state-of-the-art for inferencing 3D dense segmentation with only sparse annotations. Particularly, with $5\%$ annotations of raw data, GIDSeg outperforms other 3D segmentation methods.

cs.CV↗

arXiv2021

A Novel Framework Integrating AI Model and Enzymological Experiments Promotes Identification of SARS-CoV-2 3CL Protease Inhibitors and Activity-based Probe

The identification of protein-ligand interaction plays a key role in biochemical research and drug discovery. Although deep learning has recently shown great promise in discovering new drugs, there remains a gap between deep learning-based and experimental approaches. Here we propose a novel framework, named AIMEE, integrating AI Model and Enzymology Experiments, to identify inhibitors against 3CL protease of SARS-CoV-2, which has taken a significant toll on people across the globe. From a bioactive chemical library, we have conducted two rounds of experiments and identified six novel inhibitors with a hit rate of 29.41%, and four of them showed an IC50 value less than 3 μM. Moreover, we explored the interpretability of the central model in AIMEE, mapping the deep learning extracted features to domain knowledge of chemical properties. Based on this knowledge, a commercially available compound was selected and proven to be an activity-based probe of 3CLpro. This work highlights the great potential of combining deep learning models and biochemical experiments for intelligent iteration and expanding the boundaries of drug discovery.

q-bio.MN↗

arXiv2021

Improving the Precision of Optical Metrology by Detecting Fewer Photons

In optical metrological protocols to measure physical quantities, it is, in principle, always beneficial to increase photon number to improve measurement precision. However, practical constraints prevent arbitrary increase of n due to the imperfections of a practical detector, especially when the detector response is dominated by saturation effect. In this work, we show that a modified weak measurement protocol, namely, biased weak measurement significantly improves the precision of optical metrology in the presence of saturation effect. This method detects an ultra-small fraction of photons while maintains considerable amount of metrological information. The biased pre-coupling leads to an additional reduction of photons in the post-selection and generates an extinction point in the spectrum distribution, which is extremely sensitive to the estimated parameter and difficult to be saturated. Therefore, the Fisher information can be persistently enhanced by increasing the photon number. In our magnetic-sensing experiment, biased weak measurement achieves precision approximately one order of magnitude better than those of previously used methods. The proposed method can be applied in various optical measurement schemes to circumvent detector saturation effect with low-cost apparatuses.

quant-ph↗

arXiv2020

Taming nucleon density distributions with deep neural network method

We investigate the density distributions of finite nuclei employing a well-designed deep neural network method. We calculate the target nucleon density distributions with Skyrme density functional theories, which are used to train the networks. We find that the training with only about $10\%$ nuclei ($300-400$) is sufficient to describe the nucleon density distributions of all the nuclear chart within 2\% relative error. The relative error comes to 5\% when about 200 proton(neutron) density distributions are used for training. We obtained very similar results for different Skyrme density functional theories. Therefore the ability to train networks is weakly dependent on the theoretical model. Moreover, in the process of machine learning, there is a turning point showing the transition from the Fermi-like distribution to the realistic Skyrme distribution, which provides significant properties of convergence process.

nucl-th↗

arXiv2020

Classical Communication Enhanced Quantum State Verification

Quantum state verification provides an efficient approach to characterize the reliability of quantum devices for generating certain target states. The figure of merit of a specific strategy is the estimated infidelity $ε$ of the tested state to the target state, given a certain number of performed measurements n. Entangled measurements constitute the globally optimal strategy and achieve the scaling that εis inversely proportional to n. Recent advances show that it is possible to achieve the same scaling simply with non-adaptive local measurements, however, the performance is still worse than the globally optimal bound up to a constant factor. In this work, by introducing classical communication, we experimentally implement an adaptive quantum state verification. The constant-factor is minimized from ~2.5 to 1.5 in this experiment, which means that only 60% measurements are required to achieve a certain value of εcompared to optimal non-adaptive local strategy. Our results indicate that classical communication significantly enhances the performance of quantum state verification, and leads to an efficiency that further approaches the globally optimal bound.

quant-ph↗

arXiv2019

FusionMapping: Learning Depth Prediction with Monocular Images and 2D Laser Scans

Acquiring accurate three-dimensional depth information conventionally requires expensive multibeam LiDAR devices. Recently, researchers have developed a less expensive option by predicting depth information from two-dimensional color imagery. However, there still exists a substantial gap in accuracy between depth information estimated from two-dimensional images and real LiDAR point-cloud. In this paper, we introduce a fusion-based depth prediction method, called FusionMapping. This is the first method that fuses colored imagery and two-dimensional laser scan to estimate depth in-formation. More specifically, we propose an autoencoder-based depth prediction network and a novel point-cloud refinement network for depth estimation. We analyze the performance of our FusionMapping approach on the KITTI LiDAR odometry dataset and an indoor mobile robot system. The results show that our introduced approach estimates depth with better accuracy when compared to existing methods.

cs.CV↗

arXiv2019

LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment Analysis

Point cloud based place recognition is still an open issue due to the difficulty in extracting local features from the raw 3D point cloud and generating the global descriptor, and it's even harder in the large-scale dynamic environments. In this paper, we develop a novel deep neural network, named LPD-Net (Large-scale Place Description Network), which can extract discriminative and generalizable global descriptors from the raw 3D point cloud. Two modules, the adaptive local feature extraction module and the graph-based neighborhood aggregation module, are proposed, which contribute to extract the local structures and reveal the spatial distribution of local features in the large-scale point cloud, with an end-to-end manner. We implement the proposed global descriptor in solving point cloud based retrieval tasks to achieve the large-scale place recognition. Comparison results show that our LPD-Net is much better than PointNetVLAD and reaches the state-of-the-art. We also compare our LPD-Net with the vision-based solutions to show the robustness of our approach to different weather and light conditions.

cs.CV↗

arXiv2019

Experimental demonstration of secure quantum remote sensing

Quantum metrology aims to enhance the precision of various measurement tasks by taking advantages of quantum properties. In many scenarios, precision is not the sole target; the acquired information must be protected once it is generated in the sensing process. Considering a remote sensing scenario where a local site performs cooperative sensing with a remote site to collect private information at the remote site, the loss of sensing data inevitably causes private information to be revealed. Quantum key distribution is known to be a reliable solution for secure data transmission, however, it fails if an eavesdropper accesses the sensing data generated at a remote site. In this study, we demonstrate that by sharing entanglement between local and remote sites, secure quantum remote sensing can be realized, and the secure level is characterized by asymmetric Fisher information gain. Concretely, only the local site can acquire the estimated parameter accurately with Fisher information approaching 1. In contrast, the accessible Fisher information for an eavesdropper is nearly zero even if he/she obtains the raw sensing data at the remote site. This achievement is primarily due to the nonlocal calibration and steering of the probe state at the remote site. Our results explore one significant advantage of ``quantumness'' and extend the notion of quantum metrology to the security realm.

quant-ph↗

arXiv2019

Population stability risks and biophysical benefits of cell-cell fusion in macrophage, osteoclast, and giant multinucleated cells

Plant and animal cells are commonly understood as acquiring specialized functions through differentiation and asymmetric division. However, unique capabilities are also acquired when two or more cells fuse together, mixing cytoplasmic and genetic material. Here, we combine imaging experiments with biophysical modeling to perform the first risk-benefit analysis of cell-cell fusion. On one hand, we find fusion introduces an intrinsic instability to the population dynamics. On the other hand, we measure an unusual physiological scaling suggesting these cells grow substantially larger at lower energetic costs. Further analysis of the cytoskeleton finds a size-associated phase separation of F-actin that self-organizes multinucleated cell phenotypes.

physics.bio-ph↗

arXiv2019

Experimental Optimal Verification of Entangled States using Local Measurements

The initialization of a quantum system into a certain state is a crucial aspect of quantum information science. While a variety of measurement strategies have been developed to characterize how well the system is initialized, for a given one, there is in general a trade-off between its efficiency and the accessible information of the quantum state. Conventional quantum state tomography can characterize unknown states by reconstructing the density matrix; however, its exponentially expensive postprocessing is likely to produce a deviate result. Alternatively, quantum state verification provides a technique to quantify the prepared state with significantly fewer measurements, especially for quantum entangled states. Here, we experimentally implement an optimal verification of entangled states with local measurements, where the estimated infidelity is inversely proportional to the number of measurements. The utilized strategy is tolerant of the impurity of realistic states, hence being highly robust in a practical sense. Even more valuable, our method only requires local measurements, which incurs only a small constant-factor (<2.5) penalty compared to the globally optimal strategy requiring nonlocal measurements.

quant-ph↗

arXiv2019

A Multi-Domain Feature Learning Method for Visual Place Recognition

Visual Place Recognition (VPR) is an important component in both computer vision and robotics applications, thanks to its ability to determine whether a place has been visited and where specifically. A major challenge in VPR is to handle changes of environmental conditions including weather, season and illumination. Most VPR methods try to improve the place recognition performance by ignoring the environmental factors, leading to decreased accuracy decreases when environmental conditions change significantly, such as day versus night. To this end, we propose an end-to-end conditional visual place recognition method. Specifically, we introduce the multi-domain feature learning method (MDFL) to capture multiple attribute-descriptions for a given place, and then use a feature detaching module to separate the environmental condition-related features from those that are not. The only label required within this feature learning pipeline is the environmental condition. Evaluation of the proposed method is conducted on the multi-season \textit{NORDLAND} dataset, and the multi-weather \textit{GTAV} dataset. Experimental results show that our method improves the feature robustness against variant environmental conditions.

cs.RO↗

arXiv2019

MRS-VPR: a multi-resolution sampling based global visual place recognition method

Place recognition and loop closure detection are challenging for long-term visual navigation tasks. SeqSLAM is considered to be one of the most successful approaches to achieving long-term localization under varying environmental conditions and changing viewpoints. It depends on a brute-force, time-consuming sequential matching method. We propose MRS-VPR, a multi-resolution, sampling-based place recognition method, which can significantly improve the matching efficiency and accuracy in sequential matching. The novelty of this method lies in the coarse-to-fine searching pipeline and a particle filter-based global sampling scheme, that can balance the matching efficiency and accuracy in the long-term navigation task. Moreover, our model works much better than SeqSLAM when the testing sequence has a much smaller scale than the reference sequence. Our experiments demonstrate that the proposed method is efficient in locating short temporary trajectories within long-term reference ones without losing accuracy compared to SeqSLAM.

cs.RO↗