arXiv subjects

Long Chen

Publications and source records attributed to Long Chen.

At least 523 records · Page 29Linked to original sources

arXiv2021

Probing the tau electric dipole moment at BEPC-II collider energies

We investigate the prospects of the search for a nonzero $τ$ EDM form factor $d_τ(s)$ in $τ$ pair production by $e^+ e^-$ collisions at BEPC-II collider energies. We compute the expectation values and covariances of simple and optimal $CP$-odd observables for $τ$-pair production at $\sqrt{s}=4.18$ GeV and $4.95$ GeV with subsequent decays of $τ^\pm$ into major leptonic or semihadronic modes. For the $τ$ decays to two pions and three charged pions we take the full kinematic information of the hadronic system into account. Applying cuts and using realistic assumptions on the eventually attainable integrated luminosities at these energies, ${\cal L}(4.18) = 3\times 10^4~{\rm pb}^{-1}$ and ${\cal L}(4.95) = 10^4~{\rm pb}^{-1}$, respectively, we find the following. By taking into account purely semihadronic and semihadronic-leptonic $τ^+τ^-$ decays one can achieve with optimal $CP$-odd observables the 1 s.d. sensitivities $δ{\rm Re} d_τ= 4.5\times 10^{-18}$ ecm ($5.3\times 10^{-18}$ ecm) and $δ{\rm Im} d_τ= 2.2 \times 10^{-18}$ ecm ($2.7\times 10^{-18}$ ecm) at $\sqrt{s}=4.18$ GeV ($4.95$ GeV).

hep-ph↗

arXiv2021

Video Relation Detection via Tracklet based Visual Transformer

Video Visual Relation Detection (VidVRD), has received significant attention of our community over recent years. In this paper, we apply the state-of-the-art video object tracklet detection pipeline MEGA and deepSORT to generate tracklet proposals. Then we perform VidVRD in a tracklet-based manner without any pre-cutting operations. Specifically, we design a tracklet-based visual Transformer. It contains a temporal-aware decoder which performs feature interactions between the tracklets and learnable predicate query embeddings, and finally predicts the relations. Experimental results strongly demonstrate the superiority of our method, which outperforms other methods by a large margin on the Video Relation Understanding (VRU) Grand Challenge in ACM Multimedia 2021. Codes are released at https://github.com/Dawn-LX/VidVRD-tracklets.

cs.CV↗

arXiv2021

FMMformer: Efficient and Flexible Transformer via Decomposed Near-field and Far-field Attention

We propose FMMformers, a class of efficient and flexible transformers inspired by the celebrated fast multipole method (FMM) for accelerating interacting particle simulation. FMM decomposes particle-particle interaction into near-field and far-field components and then performs direct and coarse-grained computation, respectively. Similarly, FMMformers decompose the attention into near-field and far-field attention, modeling the near-field attention by a banded matrix and the far-field attention by a low-rank matrix. Computing the attention matrix for FMMformers requires linear complexity in computational time and memory footprint with respect to the sequence length. In contrast, standard transformers suffer from quadratic complexity. We analyze and validate the advantage of FMMformers over the standard transformer on the Long Range Arena and language modeling benchmarks. FMMformers can even outperform the standard transformer in terms of accuracy by a significant margin. For instance, FMMformers achieve an average classification accuracy of $60.74\%$ over the five Long Range Arena tasks, which is significantly better than the standard transformer's average accuracy of $58.70\%$.

cs.LG↗

arXiv2021

Structured Context Enhancement Network for Mouse Pose Estimation

Automated analysis of mouse behaviours is crucial for many applications in neuroscience. However, quantifying mouse behaviours from videos or images remains a challenging problem, where pose estimation plays an important role in describing mouse behaviours. Although deep learning based methods have made promising advances in human pose estimation, they cannot be directly applied to pose estimation of mice due to different physiological natures. Particularly, since mouse body is highly deformable, it is a challenge to accurately locate different keypoints on the mouse body. In this paper, we propose a novel Hourglass network based model, namely Graphical Model based Structured Context Enhancement Network (GM-SCENet) where two effective modules, i.e., Structured Context Mixer (SCM) and Cascaded Multi-Level Supervision (CMLS) are subsequently implemented. SCM can adaptively learn and enhance the proposed structured context information of each mouse part by a novel graphical model that takes into account the motion difference between body parts. Then, the CMLS module is designed to jointly train the proposed SCM and the Hourglass network by generating multi-level information, increasing the robustness of the whole network.Using the multi-level prediction information from SCM and CMLS, we develop an inference method to ensure the accuracy of the localisation results. Finally, we evaluate our proposed approach against several baselines...

cs.CV↗

arXiv2021

A Unified Convergence Analysis of First Order Convex Optimization Methods via Strong Lyapunov Functions

We present a unified convergence analysis for first order convex optimization methods using the concept of strong Lyapunov conditions. Combining this with suitable time scaling factors, we are able to handle both convex and strong convex cases, and establish contraction properties of Lyapunov functions for many existing ordinary differential equation models. Then we derive prevailing first order optimization algorithms, such as proximal gradient methods, heavy ball methods (also known as momentum methods), Nesterov accelerated gradient methods, and accelerated proximal gradient methods from numerical discretizations of corresponding dynamical systems. We also apply strong Lyapunov conditions to the discrete level and provide a more systematical analysis framework. Another contribution is a novel second order dynamical system called Hessian-driven Nesterov accelerated gradient flow which can be used to design and analyze accelerated first order methods for smooth and non-smooth convex optimizations.

math.OC↗

arXiv2021

Cross-View Tracking for Multi-Human 3D Pose Estimation at over 100 FPS

Estimating 3D poses of multiple humans in real-time is a classic but still challenging task in computer vision. Its major difficulty lies in the ambiguity in cross-view association of 2D poses and the huge state space when there are multiple people in multiple views. In this paper, we present a novel solution for multi-human 3D pose estimation from multiple calibrated camera views. It takes 2D poses in different camera coordinates as inputs and aims for the accurate 3D poses in the global coordinate. Unlike previous methods that associate 2D poses among all pairs of views from scratch at every frame, we exploit the temporal consistency in videos to match the 2D inputs with 3D poses directly in 3-space. More specifically, we propose to retain the 3D pose for each person and update them iteratively via the cross-view multi-human tracking. This novel formulation improves both accuracy and efficiency, as we demonstrated on widely-used public datasets. To further verify the scalability of our method, we propose a new large-scale multi-human dataset with 12 to 28 camera views. Without bells and whistles, our solution achieves 154 FPS on 12 cameras and 34 FPS on 28 cameras, indicating its ability to handle large-scale real-world applications. The proposed dataset is released at https://github.com/longcw/crossview_3d_pose_tracking.

cs.CV↗

arXiv2021

Transductive Zero-Shot Hashing for Multilabel Image Retrieval

Hash coding has been widely used in approximate nearest neighbor search for large-scale image retrieval. Given semantic annotations such as class labels and pairwise similarities of the training data, hashing methods can learn and generate effective and compact binary codes. While some newly introduced images may contain undefined semantic labels, which we call unseen images, zeor-shot hashing techniques have been studied. However, existing zeor-shot hashing methods focus on the retrieval of single-label images, and cannot handle multi-label images. In this paper, for the first time, a novel transductive zero-shot hashing method is proposed for multi-label unseen image retrieval. In order to predict the labels of the unseen/target data, a visual-semantic bridge is built via instance-concept coherence ranking on the seen/source data. Then, pairwise similarity loss and focal quantization loss are constructed for training a hashing model using both the seen/source and unseen/target data. Extensive evaluations on three popular multi-label datasets demonstrate that, the proposed hashing method achieves significantly better results than the competing methods.

cs.CV↗

arXiv2021

A finite element elasticity complex in three dimensions

A finite element elasticity complex on tetrahedral meshes is devised. The $H^1$ conforming finite element is the smooth finite element developed by Neilan for the velocity field in a discrete Stokes complex. The symmetric div-conforming finite element is the Hu-Zhang element for stress tensors. The construction of an $H(\textrm{inc})$-conforming finite element for symmetric tensors is the main focus of this paper. The key tools of the construction are the decomposition of polynomial tensor spaces and the characterization of the trace of the $\textrm{inc}$ operator. The polynomial elasticity complex and Koszul elasticity complex are created to derive the decomposition of polynomial tensor spaces. The trace of the $\textrm{inc}$ operator is induced from a Green's identity. Trace complexes and bubble complexes are also derived to facilitate the construction. Our construction appears to be the first $H(\textrm{inc})$-conforming finite elements on tetrahedral meshes without further splits.

math.NA↗

arXiv2021

Accelerate CNNs from Three Dimensions: A Comprehensive Pruning Framework

Most neural network pruning methods, such as filter-level and layer-level prunings, prune the network model along one dimension (depth, width, or resolution) solely to meet a computational budget. However, such a pruning policy often leads to excessive reduction of that dimension, thus inducing a huge accuracy loss. To alleviate this issue, we argue that pruning should be conducted along three dimensions comprehensively. For this purpose, our pruning framework formulates pruning as an optimization problem. Specifically, it first casts the relationships between a certain model's accuracy and depth/width/resolution into a polynomial regression and then maximizes the polynomial to acquire the optimal values for the three dimensions. Finally, the model is pruned along the three optimal dimensions accordingly. In this framework, since collecting too much data for training the regression is very time-costly, we propose two approaches to lower the cost: 1) specializing the polynomial to ensure an accurate regression even with less training data; 2) employing iterative pruning and fine-tuning to collect the data faster. Extensive experiments show that our proposed algorithm surpasses state-of-the-art pruning algorithms and even neural architecture search-based algorithms.

cs.CV↗

arXiv2021

Graph-based Label Propagation for Semi-Supervised Speaker Identification

Speaker identification in the household scenario (e.g., for smart speakers) is typically based on only a few enrollment utterances but a much larger set of unlabeled data, suggesting semisupervised learning to improve speaker profiles. We propose a graph-based semi-supervised learning approach for speaker identification in the household scenario, to leverage the unlabeled speech samples. In contrast to most of the works in speaker recognition that focus on speaker-discriminative embeddings, this work focuses on speaker label inference (scoring). Given a pre-trained embedding extractor, graph-based learning allows us to integrate information about both labeled and unlabeled utterances. Considering each utterance as a graph node, we represent pairwise utterance similarity scores as edge weights. Graphs are constructed per household, and speaker identities are propagated to unlabeled nodes to optimize a global consistency criterion. We show in experiments on the VoxCeleb dataset that this approach makes effective use of unlabeled data and improves speaker identification accuracy compared to two state-of-the-art scoring methods as well as their semi-supervised variants based on pseudo-labels.

cs.SD↗

arXiv2021

SDNet: mutil-branch for single image deraining using swin

Rain streaks degrade the image quality and seriously affect the performance of subsequent computer vision tasks, such as autonomous driving, social security, etc. Therefore, removing rain streaks from a given rainy images is of great significance. Convolutional neural networks(CNN) have been widely used in image deraining tasks, however, the local computational characteristics of convolutional operations limit the development of image deraining tasks. Recently, the popular transformer has global computational features that can further facilitate the development of image deraining tasks. In this paper, we introduce Swin-transformer into the field of image deraining for the first time to study the performance and potential of Swin-transformer in the field of image deraining. Specifically, we improve the basic module of Swin-transformer and design a three-branch model to implement single-image rain removal. The former implements the basic rain pattern feature extraction, while the latter fuses different features to further extract and process the image features. In addition, we employ a jump connection to fuse deep features and shallow features. In terms of experiments, the existing public dataset suffers from image duplication and relatively homogeneous background. So we propose a new dataset Rain3000 to validate our model. Therefore, we propose a new dataset Rain3000 for validating our model. Experimental results on the publicly available datasets Rain100L, Rain100H and our dataset Rain3000 show that our proposed method has performance and inference speed advantages over the current mainstream single-image rain streaks removal models.The source code will be available at https://github.com/H-tfx/SDNet.

cs.CV↗

arXiv2021

SimNet: Learning Reactive Self-driving Simulations from Real-world Observations

In this work, we present a simple end-to-end trainable machine learning system capable of realistically simulating driving experiences. This can be used for the verification of self-driving system performance without relying on expensive and time-consuming road testing. In particular, we frame the simulation problem as a Markov Process, leveraging deep neural networks to model both state distribution and transition function. These are trainable directly from the existing raw observations without the need for any handcrafting in the form of plant or kinematic models. All that is needed is a dataset of historical traffic episodes. Our formulation allows the system to construct never seen scenes that unfold realistically reacting to the self-driving car's behaviour. We train our system directly from 1,000 hours of driving logs and measure both realism, reactivity of the simulation as the two key properties of the simulation. At the same time, we apply the method to evaluate the performance of a recently proposed state-of-the-art ML planning system trained from human driving logs. We discover this planning system is prone to previously unreported causal confusion issues that are difficult to test by non-reactive simulation. To the best of our knowledge, this is the first work that directly merges highly realistic data-driven simulations with a closed-loop evaluation for self-driving vehicles. We make the data, code, and pre-trained models publicly available to further stimulate simulation development.

cs.RO↗

arXiv2021

What data do we need for training an AV motion planner?

We investigate what grade of sensor data is required for training an imitation-learning-based AV planner on human expert demonstration. Machine-learned planners are very hungry for training data, which is usually collected using vehicles equipped with the same sensors used for autonomous operation. This is costly and non-scalable. If cheaper sensors could be used for collection instead, data availability would go up, which is crucial in a field where data volume requirements are large and availability is small. We present experiments using up to 1000 hours worth of expert demonstration and find that training with 10x lower-quality data outperforms 1x AV-grade data in terms of planner performance. The important implication of this is that cheaper sensors can indeed be used. This serves to improve data access and democratize the field of imitation-based motion planning. Alongside this, we perform a sensitivity analysis of planner performance as a function of perception range, field-of-view, accuracy, and data volume, and the reason why lower-quality data still provide good planning results.

cs.RO↗

arXiv2021

Deep Learning for Weakly-Supervised Object Detection and Object Localization: A Survey

Weakly-Supervised Object Detection (WSOD) and Localization (WSOL), i.e., detecting multiple and single instances with bounding boxes in an image using image-level labels, are long-standing and challenging tasks in the CV community. With the success of deep neural networks in object detection, both WSOD and WSOL have received unprecedented attention. Hundreds of WSOD and WSOL methods and numerous techniques have been proposed in the deep learning era. To this end, in this paper, we consider WSOL is a sub-task of WSOD and provide a comprehensive survey of the recent achievements of WSOD. Specifically, we firstly describe the formulation and setting of the WSOD, including the background, challenges, basic framework. Meanwhile, we summarize and analyze all advanced techniques and training tricks for improving detection performance. Then, we introduce the widely-used datasets and evaluation metrics of WSOD. Lastly, we discuss the future directions of WSOD. We believe that these summaries can help pave a way for future research on WSOD and WSOL.

cs.CV↗

arXiv2021

Forward-backward asymmetries of the heavy quark pair production in $e^+e^-$ collisions at $\mathcal{O}(α_s^2)$

The talk is on a computational set-up for calculating the production of a massive quark-antiquark pair in electron positron collisions to order $α_s^2$ in the coupling of quantum chromodynamics (QCD) at the differential level using the antenna subtraction method. Theoretical predictions on the production of top quark pairs in the continuum, and the bottom quark pairs at the $Z$ resonance, will be discussed. In particular, we would be focusing on the order $α_s^2$ QCD corrections to the heavy quark forward-backward asymmetry (AFB) in electron positron collisions. In the case of the AFB of bottom quarks at the $Z$ resonance, the QCD corrections are determined with respect to both the bottom quark axis and the thrust axis. We will also briefly discuss improvements on these QCD corrections brought by applying the optimization procedure based on the Principle of Maximum Conformality.

hep-ph↗

arXiv2021

Renormalization of the flavor-singlet axial-vector current and its anomaly in dimensional regularization

The renormalization constant $Z_J$ of the flavor-singlet axial-vector current with a non-anticommuting $γ_5$ in dimensional regularization is determined to order $α_s^3$ in QCD with massless quarks. The result is obtained by computing the matrix elements of the operators appearing in the axial-anomaly equation $\big[\partial_μ J^μ_{5} \big]_{R} = \frac{α_s}{4 π}\, n_f\, {\displaystyle \mathrm{T}_{F}} \, \big[F \tilde{F} \big]_{R}$ between the vacuum and a state of two (off-shell) gluons to 4-loop order. Furthermore, through this computation, the conjectured equality between the $\overline{\mathrm{MS}}$ renormalization constant $Z_{F\tilde{F}}$ associated with the operator $\big[F \tilde{F} \big]_{R}$ and that of $α_s$ is verified explicitly to hold true at 4-loop order. This equality automatically ensures a relation between the respective anomalous dimensions, $γ_{\scriptscriptstyle J} = \frac{α_s}{4 π}\, n_f\, {\displaystyle \mathrm{T}_{F}} \, γ_{\scriptscriptstyle FJ} $, at order $α_s^4$ given the validity of the axial-anomaly equation which was used to determine the non-$\overline{\mathrm{MS}}$ piece of $Z_J$ for the particular $γ_5$ prescription in use.

hep-ph↗

arXiv2021

A prescription for projectors to compute helicity amplitudes in D dimensions

This article discusses a prescription to compute polarized dimensionally regularized amplitudes, providing a recipe for constructing simple and general polarized amplitude projectors in D dimensions that avoids conventional Lorentz tensor decomposition and avoids also dimensional splitting. Because of the latter, commutation between Lorentz index contraction and loop integration is preserved within this prescription, which entails certain technical advantages. The usage of these D-dimensional polarized amplitude projectors results in helicity amplitudes that can be expressed solely in terms of external momenta, but different from those defined in the existing dimensional regularization schemes. Furthermore, we argue that despite being different from the conventional dimensional regularization scheme (CDR), owing to the amplitude-level factorization of ultraviolet and infrared singularities, our prescription can be used, within an infrared subtraction framework, in a hybrid way without re-calculating the (process-independent) integrated subtraction coefficients, many of which are available in CDR. This hybrid CDR-compatible prescription is shown to be unitary. We include two examples to demonstrate this explicitly and also to illustrate its usage in practice.

hep-ph↗

arXiv2021

The electric dipole moment of the tau lepton revisited

We reconsider the issue of the search for a nonzero electric dipole form factor (EDM) $d_τ(s)$ using optimal observables in $τ^+τ^-$ production by $e^+ e^-$ collisions in the center-of-mass energy range from the $τ$-pair threshold to about $\sqrt{s} \sim 15$ GeV. We discuss the general formalism of optimal observables and apply it to two $CP$-odd observables that are sensitive to the real and imaginary part of $d_τ(s)$, respectively. We compute the expectation values and covariances of these optimal $CP$ observables for $τ$-pair production at $\sqrt{s}=10.58$ GeV with subsequent decays of $τ^\pm$ into major leptonic or semihadronic modes. For the $τ$ decays to two pions and three charged pions we take the full kinematic information of the hadronic system into account. Assuming that the Belle II experiment at the KEKB accelerator will eventually analyze data corresponding to an integrated luminosity of 50 ab$^{-1}$ and applying acceptance cuts on the final-state pions we find that 1~s.d. sensitivities $δ{\rm Re} d_τ= 6.8 \times 10^{-20}$ e cm and $δ{\rm} Im d_τ= 4.0 \times 10^{-20}$ e cm can be obtained with events where both $τ$'s decay semihadronically. In the ideal case where no cuts on the final-state particles are applied we find with 50 ab$^{-1}$ at $\sqrt{s}=10.58$ GeV corresponding to $4.5 \times 10^{10}$ $τ^+ τ^-$ events the 1 s.d. sensitivities $δ{\rm Re} d_τ= 5.8 \times 10^{-20}$ e cm and $δ{\rm} Im d_τ= 3.2 \times 10^{-20}$ e cm, again for events where both $τ$ leptons decay semihadronically. Furthermore, we analyze the potential magnitude of the $τ$ EDM form factor in the type-II two-Higgs doublet extension and in two scalar leptoquark extensions of the Standard Model.

hep-ph↗