arXiv subjects

Liang Zheng

Publications and source records attributed to Liang Zheng.

At least 145 records · Page 8Linked to original sources

arXiv2021

The 5th AI City Challenge

The AI City Challenge was created with two goals in mind: (1) pushing the boundaries of research and development in intelligent video analysis for smarter cities use cases, and (2) assessing tasks where the level of performance is enough to cause real-world adoption. Transportation is a segment ripe for such adoption. The fifth AI City Challenge attracted 305 participating teams across 38 countries, who leveraged city-scale real traffic data and high-quality synthetic data to compete in five challenge tracks. Track 1 addressed video-based automatic vehicle counting, where the evaluation being conducted on both algorithmic effectiveness and computational efficiency. Track 2 addressed city-scale vehicle re-identification with augmented synthetic data to substantially increase the training set for the task. Track 3 addressed city-scale multi-target multi-camera vehicle tracking. Track 4 addressed traffic anomaly detection. Track 5 was a new track addressing vehicle retrieval using natural language descriptions. The evaluation system shows a general leader board of all submitted results, and a public leader board of results limited to the contest participation rules, where teams are not allowed to use external data in their work. The public leader board shows results more close to real-world situations where annotated data is limited. Results show the promise of AI in Smarter Transportation. State-of-the-art performance for some tasks shows that these technologies are ready for adoption in real-world systems.

cs.CV↗

arXiv2021

VTNet: Visual Transformer Network for Object Goal Navigation

Object goal navigation aims to steer an agent towards a target object based on observations of the agent. It is of pivotal importance to design effective visual representations of the observed scene in determining navigation actions. In this paper, we introduce a Visual Transformer Network (VTNet) for learning informative visual representation in navigation. VTNet is a highly effective structure that embodies two key properties for visual representations: First, the relationships among all the object instances in a scene are exploited; Second, the spatial locations of objects and image regions are emphasized so that directional navigation signals can be learned. Furthermore, we also develop a pre-training scheme to associate the visual representations with navigation signals, and thus facilitate navigation policy learning. In a nutshell, VTNet embeds object and region features with their location cues as spatial-aware descriptors and then incorporates all the encoded descriptors through attention operations to achieve informative representation for navigation. Given such visual representations, agents are able to explore the correlations between visual observations and navigation actions. For example, an agent would prioritize "turning right" over "turning left" when the visual representation emphasizes on the right side of activation map. Experiments in the artificial environment AI2-Thor demonstrate that VTNet significantly outperforms state-of-the-art methods in unseen testing environments.

cs.CV↗

arXiv2021

Joint Discriminative and Generative Learning for Person Re-identification

Person re-identification (re-id) remains challenging due to significant intra-class variations across different cameras. Recently, there has been a growing interest in using generative models to augment training data and enhance the invariance to input changes. The generative pipelines in existing methods, however, stay relatively separate from the discriminative re-id learning stages. Accordingly, re-id models are often trained in a straightforward manner on the generated data. In this paper, we seek to improve learned re-id embeddings by better leveraging the generated data. To this end, we propose a joint learning framework that couples re-id learning and data generation end-to-end. Our model involves a generative module that separately encodes each person into an appearance code and a structure code, and a discriminative module that shares the appearance encoder with the generative module. By switching the appearance or structure codes, the generative module is able to generate high-quality cross-id composed images, which are online fed back to the appearance encoder and used to improve the discriminative module. The proposed joint learning framework renders significant improvement over the baseline without using generated data, leading to the state-of-the-art performance on several benchmark datasets.

cs.CV↗

arXiv2021

Source Free Domain Adaptation with Image Translation

Effort in releasing large-scale datasets may be compromised by privacy and intellectual property considerations. A feasible alternative is to release pre-trained models instead. While these models are strong on their original task (source domain), their performance might degrade significantly when deployed directly in a new environment (target domain), which might not contain labels for training under realistic settings. Domain adaptation (DA) is a known solution to the domain gap problem, but usually requires labeled source data. In this paper, we study the problem of source free domain adaptation (SFDA), whose distinctive feature is that the source domain only provides a pre-trained model, but no source data. Being source free adds significant challenges to DA, especially when considering that the target dataset is unlabeled. To solve the SFDA problem, we propose an image translation approach that transfers the style of target images to that of unseen source images. To this end, we align the batch-wise feature statistics of generated images to that stored in batch normalization layers of the pre-trained model. Compared with directly classifying target images, higher accuracy is obtained with these style transferred images using the pre-trained model. On several image classification datasets, we show that the above-mentioned improvements are consistent and statistically significant.

cs.CV↗

arXiv2021

Learning to Structure an Image with Few Colors

Color and structure are the two pillars that construct an image. Usually, the structure is well expressed through a rich spectrum of colors, allowing objects in an image to be recognized by neural networks. However, under extreme limitations of color space, the structure tends to vanish, and thus a neural network might fail to understand the image. Interested in exploring this interplay between color and structure, we study the scientific problem of identifying and preserving the most informative image structures while constraining the color space to just a few bits, such that the resulting image can be recognized with possibly high accuracy. To this end, we propose a color quantization network, ColorCNN, which learns to structure the images from the classification loss in an end-to-end manner. Given a color space size, ColorCNN quantizes colors in the original image by generating a color index map and an RGB color palette. Then, this color-quantized image is fed to a pre-trained task network to evaluate its performance. In our experiment, with only a 1-bit color space (i.e., two colors), the proposed network achieves 82.1% top-1 accuracy on the CIFAR10 dataset, outperforming traditional color quantization methods by a large margin. For applications, when encoded with PNG, the proposed color quantization shows superiority over other image compression methods in the extremely low bit-rate regime. The code is available at: https://github.com/hou-yz/color_distillation.

cs.CV↗

arXiv2021

Boosting Semi-Supervised Face Recognition with Noise Robustness

Although deep face recognition benefits significantly from large-scale training data, a current bottleneck is the labelling cost. A feasible solution to this problem is semi-supervised learning, exploiting a small portion of labelled data and large amounts of unlabelled data. The major challenge, however, is the accumulated label errors through auto-labelling, compromising the training. This paper presents an effective solution to semi-supervised face recognition that is robust to the label noise aroused by the auto-labelling. Specifically, we introduce a multi-agent method, named GroupNet (GN), to endow our solution with the ability to identify the wrongly labelled samples and preserve the clean samples. We show that GN alone achieves the leading accuracy in traditional supervised face recognition even when the noisy labels take over 50\% of the training data. Further, we develop a semi-supervised face recognition solution, named Noise Robust Learning-Labelling (NRoLL), which is based on the robust training ability empowered by GN. It starts with a small amount of labelled data and consequently conducts high-confidence labelling on a large amount of unlabelled data to boost further training. The more data is labelled by NRoLL, the higher confidence is with the label in the dataset. To evaluate the competitiveness of our method, we run NRoLL with a rough condition that only one-fifth of the labelled MSCeleb is available and the rest is used as unlabelled data. On a wide range of benchmarks, our method compares favorably against the state-of-the-art methods.

cs.CV↗

arXiv2021

Multiview Detection with Feature Perspective Transformation

Incorporating multiple camera views for detection alleviates the impact of occlusions in crowded scenes. In a multiview system, we need to answer two important questions when dealing with ambiguities that arise from occlusions. First, how should we aggregate cues from the multiple views? Second, how should we aggregate unreliable 2D and 3D spatial information that has been tainted by occlusions? To address these questions, we propose a novel multiview detection system, MVDet. For multiview aggregation, existing methods combine anchor box features from the image plane, which potentially limits performance due to inaccurate anchor box shapes and sizes. In contrast, we take an anchor-free approach to aggregate multiview information by projecting feature maps onto the ground plane (bird's eye view). To resolve any remaining spatial ambiguity, we apply large kernel convolutions on the ground plane feature map and infer locations from detection peaks. Our entire model is end-to-end learnable and achieves 88.2% MODA on the standard Wildtrack dataset, outperforming the state-of-the-art by 14.1%. We also provide detailed analysis of MVDet on a newly introduced synthetic dataset, MultiviewX, which allows us to control the level of occlusion. Code and MultiviewX dataset are available at https://github.com/hou-yz/MVDet.

cs.CV↗

arXiv2021

Visualizing Adapted Knowledge in Domain Transfer

A source model trained on source data and a target model learned through unsupervised domain adaptation (UDA) usually encode different knowledge. To understand the adaptation process, we portray their knowledge difference with image translation. Specifically, we feed a translated image and its original version to the two models respectively, formulating two branches. Through updating the translated image, we force similar outputs from the two branches. When such requirements are met, differences between the two images can compensate for and hence represent the knowledge difference between models. To enforce similar outputs from the two branches and depict the adapted knowledge, we propose a source-free image translation method that generates source-style images using only target images and the two models. We visualize the adapted knowledge on several datasets with different UDA methods and find that generated images successfully capture the style difference between the two domains. For application, we show that generated images enable further tuning of the target model without accessing source data. Code available at https://github.com/hou-yz/DA_visualization.

cs.CV↗

arXiv2021

Exploring the Collective Phenomenon at the Electron-Ion Collider

Based on rare fluctuations in strong interactions, we argue that there is a strong physical resemblance between the high multiplicity events in photo-nuclear collisions and those in $pA$ collisions, in which interesting long range collective phenomena are discovered. This indicates that the collectivity can also be studied in certain kinematic region of the upcoming Electron-Ion Collider (EIC) where the incoming virtual photon has a sufficiently long lifetime. Using a model in the Color Glass Condensate formalism, we first show that the initial state interactions can explain the recent ATLAS azimuthal correlation results measured in the photo-nuclear collisions, and then we provide quantitative predictions for the long range correlations in $eA$ collisions in the EIC regime. With the unprecedented precision and the ability to change the size of the collisional system, the high luminosity EIC will open a new window to explore the physical mechanism responsible for the collective phenomenon.

hep-ph↗

arXiv2021

Using local nuclear scaling of initial condition parameters to improve the system size dependence of transport model descriptions of nuclear collisions

We extensively study the system size dependence of nuclear collisions with a multi-phase transport model. Previously certain key parameters for the initial condition needed significantly different values for $pp$ and central $AA$ collisions for the model to reasonably describe the yields and transverse momentum spectra of the bulk matter in those collision systems. Here we scale two key parameters, the Lund string fragmentation parameter $b_L$ and the minijet transverse momentum cutoff $p_0$, with local nuclear thickness functions from the two colliding nuclei. This allows the model to use the parameter values for $pp$ collisions with the local nuclear scaling to describe the system size and centrality dependences of nuclear collisions self consistently. In addition to providing good descriptions of $pp$ collisions from 23.6 GeV to 13 TeV and reasonable descriptions of the centrality dependence of charged particle yields for Au+Au collisions from $7.7A$ GeV to $200A$ GeV and Pb+Pb collisions at LHC energies, the improved model can now well describe the centrality dependence of the mean transverse momentum of charged particles below $p_{\rm T} \lesssim 2$ GeV. It works similarly well for smaller systems including $p$Pb, Cu+Cu and Xe+Xe collisions.

nucl-th↗

arXiv2021

Charge-dependent transverse momentum and its impact on the search for the chiral magnetic wave

The chiral magnetic wave (CMW) is sought using the charge asymmetry ($A_{\rm ch}$) dependence of anisotropic flow in heavy-ion collisions. The charge dependent transverse momentum ($p_{\rm T}$), however, could play a role as a background. With the string fragmentation models, including PYTHIA, we demonstrate the origin of the $A_{\rm ch}-p_{\rm T}$ correlation and its connection with the local charge conservation (LCC). The impact of $A_{\rm ch}-p_{\rm T}$ and its behavior in varied kinematic windows are also discussed. This study provides more insights for the search for the CMW and comprehending the collective motion of the quark-gluon plasma.

nucl-th↗

arXiv2021

The study of exotic state $Z_c^{\pm}(3900)$ decaying to $J/ψπ^{\pm}$ in the $pp$ collisions at $\sqrt{s}$ = 1.96, 7, and 13 TeV

A dynamically constrained phase-space coalescence model plus PACIAE model was used to predict the exotic resonant state $Z_c^{\pm}(3900)$ yield, transverse momentum distribution, and the rapidity distribution with $|y| < 6$ and $p_T < 10$ GeV/c in $pp$ collisions at $\sqrt{s} = 1.96, 7$ and 13 TeV, respectively. The yield of the $Z_c^{\pm}(3900)$ is estimated to be around $10^{-6}$ to $10^{-5}$. We also present the energy dependence of the transverse momentum distributions and rapidity distributions for ${Z_c^{+}(3900)}$ and ${Z_c^{-}(3900)}$. The production of ${Z_c^{+}(3900)}$ and its anti-particle ${Z_c^{-}(3900)}$ is found to be quite similar to each other.

hep-ph↗

arXiv2021

Predictions for production of $\rm{^3_ΛH}$ and $\rm{{^3_{\overline Λ}\overline H}}$ in isobaric $^{96}_{44}$Ru+$^{96}_{44}$Ru and $^{96}_{40}$Zr+$^{96}_{40}$Zr collisions at $\sqrt{s_{\rm{NN}}}$ = 200 GeV

The production of $\rm{^3_ΛH}$ and $\rm{{^3_{\overline Λ}\overline H}}$, as well as $\rm{^3H}$, $\rm{^3\overline H}$, $\rm{^3He}$, and $\rm{^3\overline {He}}$ are studied in central collisions of isobars $^{96}_{44}$Ru+$^{96}_{44}$Ru and $^{96}_{40}$Zr+$^{96}_{40}$Zr at $\sqrt{s_{\rm{NN}}}=200$ GeV, using the dynamically constrained phase-space coalescence model and the {\footnotesize PACIAE} model with chiral magnetic effect. The yield, yield ratio, coalescence parameters, and strangeness population factor of (anti-)hypertriton and (anti-)nuclei produced in isobaric $^{96}_{44}$Ru+$^{96}_{44}$Ru and $^{96}_{40}$Zr+$^{96}_{40}$Zr collisions are predicted. The (anti-)hypertriton and (anti-)nuclei production is found to be insensitive to the chiral magnetic effects. Experimental data of Cu+Cu, Au+Au and Pb+Pb collisions from RHIC, LHC, and the results of {\footnotesize PACIAE+DCPC} model are presented in the results for comparison.

nucl-th↗

arXiv2021

Electron-Ion Collider in China

Lepton scattering is an established ideal tool for studying inner structure of small particles such as nucleons as well as nuclei. As a future high energy nuclear physics project, an Electron-ion collider in China (EicC) has been proposed. It will be constructed based on an upgraded heavy-ion accelerator, High Intensity heavy-ion Accelerator Facility (HIAF) which is currently under construction, together with a new electron ring. The proposed collider will provide highly polarized electrons (with a polarization of $\sim$80%) and protons (with a polarization of $\sim$70%) with variable center of mass energies from 15 to 20 GeV and the luminosity of (2-3) $\times$ 10$^{33}$ cm$^{-2}$ s$^{-1}$. Polarized deuterons and Helium-3, as well as unpolarized ion beams from Carbon to Uranium, will be also available at the EicC. The main foci of the EicC will be precision measurements of the structure of the nucleon in the sea quark region, including 3D tomography of nucleon; the partonic structure of nuclei and the parton interaction with the nuclear environment; the exotic states, especially those with heavy flavor quark contents. In addition, issues fundamental to understanding the origin of mass could be addressed by measurements of heavy quarkonia near-threshold production at the EicC. In order to achieve the above-mentioned physics goals, a hermetical detector system will be constructed with cutting-edge technologies. This document is the result of collective contributions and valuable inputs from experts across the globe. The EicC physics program complements the ongoing scientific programs at the Jefferson Laboratory and the future EIC project in the United States. The success of this project will also advance both nuclear and particle physics as well as accelerator and detector technology in China.

nucl-ex↗

arXiv2021

Sparse Attention Guided Dynamic Value Estimation for Single-Task Multi-Scene Reinforcement Learning

Training deep reinforcement learning agents on environments with multiple levels / scenes from the same task, has become essential for many applications aiming to achieve generalization and domain transfer from simulation to the real world. While such a strategy is helpful with generalization, the use of multiple scenes significantly increases the variance of samples collected for policy gradient computations. Current methods, effectively continue to view this collection of scenes as a single Markov decision process (MDP), and thus learn a scene-generic value function V(s). However, we argue that the sample variance for a multi-scene environment is best minimized by treating each scene as a distinct MDP, and then learning a joint value function V(s,M) dependent on both state s and MDP M. We further demonstrate that the true joint value function for a multi-scene environment, follows a multi-modal distribution which is not captured by traditional CNN / LSTM based critic networks. To this end, we propose a dynamic value estimation (DVE) technique, which approximates the true joint value function through a sparse attention mechanism over multiple value function hypothesis / modes. The resulting agent not only shows significant improvements in the final reward score across a range of OpenAI ProcGen environments, but also exhibits enhanced navigation efficiency and provides an implicit mechanism for unsupervised state-space skill decomposition.

cs.LG↗

arXiv2020

Fine-grained Classification via Categorical Memory Networks

Motivated by the desire to exploit patterns shared across classes, we present a simple yet effective class-specific memory module for fine-grained feature learning. The memory module stores the prototypical feature representation for each category as a moving average. We hypothesize that the combination of similarities with respect to each category is itself a useful discriminative cue. To detect these similarities, we use attention as a querying mechanism. The attention scores with respect to each class prototype are used as weights to combine prototypes via weighted sum, producing a uniquely tailored response feature representation for a given input. The original and response features are combined to produce an augmented feature for classification. We integrate our class-specific memory module into a standard convolutional neural network, yielding a Categorical Memory Network. Our memory module significantly improves accuracy over baseline CNNs, achieving competitive accuracy with state-of-the-art methods on four benchmarks, including CUB-200-2011, Stanford Cars, FGVC Aircraft, and NABirds.

cs.CV↗

arXiv2020

Coherent Optical Memory Baesd on A Laser-written On-chip Waveguide

Quantum memory is the core device for the construction of large-scale quantum networks. For scalable and convenient practical applications, integrated optical memories, especially on-chip optical memories, are crucial requirements because they can be easily integrated with other on-chip devices. Here, we report the coherent optical memory based on a type-IV waveguide fabricated on the surface of a rare-earth ion-doped crystal (i.e. $\mathrm{Eu^{3+}}$:$\mathrm{Y_2SiO_5}$). The properties of the optical transition ($\mathrm{{^7}F{_0}\rightarrow{^5}D{_0}}$) of the $\mathrm{Eu^{3+}}$ ions inside the surface waveguide are well preserved compared to those of the bulk crystal. Spin-wave atomic frequency comb storage is demonstrated inside the type-IV waveguide. The reliability of this device is confirmed by the high interference visibility of ${97\pm 1\%}$ between the retrieval pulse and the reference pulse. The developed on-chip optical memory paves the way towards integrated quantum nodes.

quant-ph↗

arXiv2020

Enhanced Scene Specificity with Sparse Dynamic Value Estimation

Multi-scene reinforcement learning involves training the RL agent across multiple scenes / levels from the same task, and has become essential for many generalization applications. However, the inclusion of multiple scenes leads to an increase in sample variance for policy gradient computations, often resulting in suboptimal performance with the direct application of traditional methods (e.g. PPO, A3C). One strategy for variance reduction is to consider each scene as a distinct Markov decision process (MDP) and learn a joint value function dependent on both state (s) and MDP (M). However, this is non-trivial as the agent is usually unaware of the underlying level at train / test times in multi-scene RL. Recently, Singh et al. [1] tried to address this by proposing a dynamic value estimation approach that models the true joint value function distribution as a Gaussian mixture model (GMM). In this paper, we argue that the error between the true scene-specific value function and the predicted dynamic estimate can be further reduced by progressively enforcing sparse cluster assignments once the agent has explored most of the state space. The resulting agents not only show significant improvements in the final reward score across a range of OpenAI ProcGen environments, but also exhibit increased navigation efficiency while completing a game level.

cs.LG↗