Search

This study investigates the development dilemma of ride-sharing services using real-world mobility datasets from nine cities and calibrated customers' price and detour elasticity. Through massive numerical experiments, this study reveals that while ride-sharing can benefit social welfare, it may also lead to a loss of revenue for transportation network companies (TNCs) or drivers compared with solo-hailing, which limits TNCs' motivation to develop ride-sharing services. Three key factors contributing to this revenue loss are identified: (1) the low successful sharing ratio for customers choosing ride-sharing in some cases, (2) the limited saved trip distance by pooling two customers, and (3) the potential revenue loss when pooling customers with significantly different trip fares. Furthermore, this study finds that the monetary benefits of carbon emission reductions from ride-sharing are not substantial enough to affect customers' choices between solo-hailing and ride-sharing. The findings provide a valuable reference for TNCs and governments. For TNCs, effective pricing strategies, such as dynamic pricing, should be designed to prevent revenue loss when introducing ride-sharing. Governments are suggested to subsidize ride-sharing services to solve the development dilemma and maintain or even increase social welfare benefits from ride-sharing, including reduced carbon emissions and improved vehicle occupancy rates.

Closed-loop simulation is crucial for end-to-end autonomous driving. Existing sensor simulation methods (e.g., NeRF and 3DGS) reconstruct driving scenes based on conditions that closely mirror training data distributions. However, these methods struggle with rendering novel trajectories, such as lane changes. Recent works have demonstrated that integrating world model knowledge alleviates these issues. Despite their efficiency, these approaches still encounter difficulties in the accurate representation of more complex maneuvers, with multi-lane shifts being a notable example. Therefore, we introduce ReconDreamer, which enhances driving scene reconstruction through incremental integration of world model knowledge. Specifically, DriveRestorer is proposed to mitigate artifacts via online restoration. This is complemented by a progressive data update strategy designed to ensure high-quality rendering for more complex maneuvers. To the best of our knowledge, ReconDreamer is the first method to effectively render in large maneuvers. Experimental results demonstrate that ReconDreamer outperforms Street Gaussians in the NTA-IoU, NTL-IoU, and FID, with relative improvements by 24.87%, 6.72%, and 29.97%. Furthermore, ReconDreamer surpasses DriveDreamer4D with PVG during large maneuver rendering, as verified by a relative improvement of 195.87% in the NTA-IoU metric and a comprehensive user study., Comment: Project Page: https://recondreamer.github.io

Parkinson's disease (PD), a degenerative disorder of the central nervous system, is commonly diagnosed using functional medical imaging techniques such as single-photon emission computed tomography (SPECT). In this study, we utilized two SPECT data sets (n = 634 and n = 202) from different hospitals to develop a model capable of accurately predicting PD stages, a multiclass classification task. We used the entire three-dimensional (3D) brain images as input and experimented with various model architectures. Initially, we treated the 3D images as sequences of two-dimensional (2D) slices and fed them sequentially into 2D convolutional neural network (CNN) models pretrained on ImageNet, averaging the outputs to obtain the final predicted stage. We also applied 3D CNN models pretrained on Kinetics-400. Additionally, we incorporated an attention mechanism to account for the varying importance of different slices in the prediction process. To further enhance model efficacy and robustness, we simultaneously trained the two data sets using weight sharing, a technique known as cotraining. Our results demonstrated that 2D models pretrained on ImageNet outperformed 3D models pretrained on Kinetics-400, and models utilizing the attention mechanism outperformed both 2D and 3D models. The cotraining technique proved effective in improving model performance when the cotraining data sets were sufficiently large., Comment: 34 pages, 7 figures, and 4 tables

Closed-loop simulation is essential for advancing end-to-end autonomous driving systems. Contemporary sensor simulation methods, such as NeRF and 3DGS, rely predominantly on conditions closely aligned with training data distributions, which are largely confined to forward-driving scenarios. Consequently, these methods face limitations when rendering complex maneuvers (e.g., lane change, acceleration, deceleration). Recent advancements in autonomous-driving world models have demonstrated the potential to generate diverse driving videos. However, these approaches remain constrained to 2D video generation, inherently lacking the spatiotemporal coherence required to capture intricacies of dynamic driving environments. In this paper, we introduce DriveDreamer4D, which enhances 4D driving scene representation leveraging world model priors. Specifically, we utilize the world model as a data machine to synthesize novel trajectory videos, where structured conditions are explicitly leveraged to control the spatial-temporal consistency of traffic elements. Besides, the cousin data training strategy is proposed to facilitate merging real and synthetic data for optimizing 4DGS. To our knowledge, DriveDreamer4D is the first to utilize video generation models for improving 4D reconstruction in driving scenarios. Experimental results reveal that DriveDreamer4D significantly enhances generation quality under novel trajectory views, achieving a relative improvement in FID by 32.1%, 46.4%, and 16.3% compared to PVG, S3Gaussian, and Deformable-GS. Moreover, DriveDreamer4D markedly enhances the spatiotemporal coherence of driving agents, which is verified by a comprehensive user study and the relative increases of 22.6%, 43.5%, and 15.6% in the NTA-IoU metric., Comment: Project Page: https://drivedreamer4d.github.io

This study explored the architecture of semantic segmentation and evaluated models that excel in polyp segmentation. We present an integrated framework that harnesses the advantages of different models to attain an optimal outcome. Specifically, in this framework, we fuse the learned features from convolutional and transformer models for prediction, thus engendering an ensemble technique to enhance model performance. Our experiments on polyp segmentation revealed that the proposed architecture surpassed other top models, exhibiting improved learning capacity and resilience. The code is available at https://github.com/HuangDLab/EnFormer., Comment: 13 pages, 3 figures, and 7 tables

General world models represent a crucial pathway toward achieving Artificial General Intelligence (AGI), serving as the cornerstone for various applications ranging from virtual environments to decision-making systems. Recently, the emergence of the Sora model has attained significant attention due to its remarkable simulation capabilities, which exhibits an incipient comprehension of physical laws. In this survey, we embark on a comprehensive exploration of the latest advancements in world models. Our analysis navigates through the forefront of generative methodologies in video generation, where world models stand as pivotal constructs facilitating the synthesis of highly realistic visual content. Additionally, we scrutinize the burgeoning field of autonomous-driving world models, meticulously delineating their indispensable role in reshaping transportation and urban mobility. Furthermore, we delve into the intricacies inherent in world models deployed within autonomous agents, shedding light on their profound significance in enabling intelligent interactions within dynamic environmental contexts. At last, we examine challenges and limitations of world models, and discuss their potential future directions. We hope this survey can serve as a foundational reference for the research community and inspire continued innovation. This survey will be regularly updated at: https://github.com/GigaAI-research/General-World-Models-Survey., Comment: This survey will be regularly updated at: https://github.com/GigaAI-research/General-World-Models-Survey

World models have demonstrated superiority in autonomous driving, particularly in the generation of multi-view driving videos. However, significant challenges still exist in generating customized driving videos. In this paper, we propose DriveDreamer-2, which builds upon the framework of DriveDreamer and incorporates a Large Language Model (LLM) to generate user-defined driving videos. Specifically, an LLM interface is initially incorporated to convert a user's query into agent trajectories. Subsequently, a HDMap, adhering to traffic regulations, is generated based on the trajectories. Ultimately, we propose the Unified Multi-View Model to enhance temporal and spatial coherence in the generated driving videos. DriveDreamer-2 is the first world model to generate customized driving videos, it can generate uncommon driving videos (e.g., vehicles abruptly cut in) in a user-friendly manner. Besides, experimental results demonstrate that the generated videos enhance the training of driving perception methods (e.g., 3D detection and tracking). Furthermore, video generation quality of DriveDreamer-2 surpasses other state-of-the-art methods, showcasing FID and FVD scores of 11.2 and 55.7, representing relative improvements of 30% and 50%., Comment: Project Page: https://drivedreamer2.github.io

Efficiently generating energetically stable crystal structures has long been a challenge in material design, primarily due to the immense arrangement of atoms in a crystal lattice. To facilitate the discovery of stable material, we present a framework for the generation of synthesizable materials, leveraging a point cloud representation to encode intricate structural information. At the heart of this framework lies the introduction of a diffusion model as its foundational pillar. To gauge the efficacy of our approach, we employ it to reconstruct input structures from our training datasets, rigorously validating its high reconstruction performance. Furthermore, we demonstrate the profound potential of Point Cloud-Based Crystal Diffusion (PCCD) by generating entirely new materials, emphasizing their synthesizability. Our research stands as a noteworthy contribution to the advancement of materials design and synthesis through the cutting-edge avenue of generative design instead of the conventional substitution or experience-based discovery., Comment: I have submitted to a journal

Recently a quantum simulator for the 2D Fermi-Hubbard model on an anisotropic triangular lattice has been realized, where both geometrical frustration and doping can be continuously tuned. Here we provide a comprehensive comparison between the magnetic correlations calculated by the two-particle self-consistent (TPSC) theory and those measured in this quantum simulator, at temperatures comparable to the spin exchange energy and Hubbard interactions comparable to the bandwidth. We find overall excellent agreements between the TPSC calculations and the measurements from the quantum simulator at all levels of frustration and doping. This is quite remarkable considering that the Hubbard model is already in the intermediate to strong coupling regime, for which very few methods yield reliable results. Our work showcases the potential of TPSC as a theoretical approach capable of providing reasonably accurate descriptions of the Hubbard model even at fairly strong interactions and when the frustration is present., Comment: 7 pages, 5 figures

This study outlines a numerical methodology aimed at rectifying the neutron scattering cross-sections of fundamental elements across a range of low neutron energies typically employed in general neutron scattering experiments. By using the experimental power law relationship governing the cross-section's dependence on neutron wavelength, we establish a mathematical connection between these two variables. Leveraging this relationship, the scheme of central moment expansion is adopted to correct the cross-sections that are applicable to general neutron wavelength distributions commonly encountered in experimental scenarios. Importantly, our proposed method eliminates the requirement for knowledge about the functional form of the distribution. Consequently, this approach offers the capability to reconstruct neutron scattering data without introducing distortions stemming from the energy-dependent cross-sections of different types of elements within materials during experimental measurements. Ultimately, this advancement facilitates a more precise interpretation and analysis of material structures based on their scattering signatures.

World models, especially in autonomous driving, are trending and drawing extensive attention due to their capacity for comprehending driving environments. The established world model holds immense potential for the generation of high-quality driving videos, and driving policies for safe maneuvering. However, a critical limitation in relevant research lies in its predominant focus on gaming environments or simulated settings, thereby lacking the representation of real-world driving scenarios. Therefore, we introduce DriveDreamer, a pioneering world model entirely derived from real-world driving scenarios. Regarding that modeling the world in intricate driving scenes entails an overwhelming search space, we propose harnessing the powerful diffusion model to construct a comprehensive representation of the complex environment. Furthermore, we introduce a two-stage training pipeline. In the initial phase, DriveDreamer acquires a deep understanding of structured traffic constraints, while the subsequent stage equips it with the ability to anticipate future states. The proposed DriveDreamer is the first world model established from real-world driving scenarios. We instantiate DriveDreamer on the challenging nuScenes benchmark, and extensive experiments verify that DriveDreamer empowers precise, controllable video generation that faithfully captures the structural constraints of real-world traffic scenarios. Additionally, DriveDreamer enables the generation of realistic and reasonable driving policies, opening avenues for interaction and practical applications., Comment: Project Page: https://drivedreamer.github.io

We are concerned with averaging theorems for $\epsilon$-small stochastic perturbations of integrable equations in $\mathbb{R}^d \times \mathbb{T}^n =\{(I,\varphi)\}$ $$ \dot I(t) =0,\quad \dot \varphi(t) = \theta(I), \qquad (1)$$ and in $\mathbb{R}^{2n} = \{v=(\mathbf{v}_1, \dots, \mathbf{v}_n), \; \mathbf{v}_j \in \mathbb{R}^2\}$, $$ \dot{\mathbf{v}}_k(t) =W_k(I) \mathbf{v}_k^\bot, \quad k=1, \dots, n, \qquad (2) $$ where $I=(I_1, \dots, I_n)$ is the vector of actions, $I_j = \frac12 \| \mathbf{v}_j\|^2$. The vector-functions $\theta$ and $W$ are locally Lipschitz and non-degenerate. Perturbations of these equations are assumed to be locally Lipschitz and such that some few first moments of the norms of their solutions are bounded uniformly in $\epsilon$, for $0\le t\le \epsilon^{-1} T$. For $I$-components of solutions for perturbations of (1) we establish their convergence in law to solutions of the corresponding averaged $I$-equations, when $0\le \tau := \epsilon t\le T$ and $\epsilon\to0$. Then we show that if the system of averaged $I$-equations is mixing, then the convergence is uniform in the slow time $\tau=\epsilon t\ge0$. Next using these results, for $\epsilon$-perturbed equations of (2) we construct well posed {\it effective stochastic equations} for $v(\tau)\in \mathbb{R}^{2n}$ (independent from $\epsilon$) such that when $\epsilon\to0$, actions of solutions of the perturbed equations of (2) with $t:= \tau/\epsilon$ converge in distribution to actions of solutions for the effective equations. Again, if the effective system is mixing, this convergence is uniform in the slow time $\tau \ge0$. We provide easy sufficient conditions on the perturbed equations which ensure that our results apply to their solutions.

Lithium niobate (LiNbO3, LN) devices play critical roles in quantum information processing. However, for special applications like quantum key distribution (QKD), the characteristics of materials and devices and their impact on practical systems must be intensively inquired. For the first time, we reveal that the photorefractive effect in LN can be utilized as a potential loophole to carry out malicious attacks by the eavesdroppers. We take a commercial LN-based variable optical attenuator as an example to demonstrate the method we named Induced-photorefractive attack (IPA) and propose two techniques to enable controllable attacks. Our results show that eavesdroppers can fulfill an efficient source-side attack by injecting an optimized irradiation beam with only several nanowatts, which is realistic when accessing commercial fiber channels. These measure and techniques can be employed for all individual and on-chip LN devices and initially explored a new security branch for system design and standardization of real-life QKD.

Bike sharing is emerging globally as an active, convenient, and sustainable mode of transportation. To plan successful bike-sharing systems (BSSs), many cities start from a small-scale pilot and gradually expand the system to cover more areas. For station-based BSSs, this means planning new stations based on existing ones over time, which requires prediction of the number of trips generated by these new stations across the whole system. Previous studies typically rely on relatively simple regression or machine learning models, which are limited in capturing complex spatial relationships. Despite the growing literature in deep learning methods for travel demand prediction, they are mostly developed for short-term prediction based on time series data, assuming no structural changes to the system. In this study, we focus on the trip generation problem for BSS expansion, and propose a graph neural network (GNN) approach to predicting the station-level demand based on multi-source urban built environment data. Specifically, it constructs multiple localized graphs centered on each target station and uses attention mechanisms to learn the correlation weights between stations. We further illustrate that the proposed approach can be regarded as a generalized spatial regression model, indicating the commonalities between spatial regression and GNNs. The model is evaluated based on realistic experiments using multi-year BSS data from New York City, and the results validate the superior performance of our approach compared to existing methods. We also demonstrate the interpretability of the model for uncovering the effects of built environment features and spatial interactions between stations, which can provide strategic guidance for BSS station location selection and capacity planning.

Weyl points exist in a fascinating topological state of matter with linear band crossings analogous to magnetic monopoles. Tremendous efforts have been devoted to investigate fermionic topological matters with Weyl points in the single-particle band dispersion. It remains elusive for realizing interaction-induced Weyl points, especially for bosons. Motivated by recent experimental progress in ultracold atoms, we propose a scheme to create Weyl points for Bogoliubov excitations of a bosonic superfluid in a three-dimensional cubic optical lattice. The unique design of the lattice leads to interaction-induced time-reversal symmetry breaking for a $p$-orbital superfluid, which in turn induces Weyl Bogoliubov excitations. Analogous to Weyl semimetals of electronic systems, the superfluid also support topologically protected edge modes due to the bulk-boundary correspondence.

In this paper, we prove the existence of super-hyperbolic orbits in four-body problem, which solves a conjecture of Marchal-Saari. We also prove the existence of noncollision singularities in the same model, which solves a conjecture of Anosov. Moreover, the two type of solutions coexist for certain mass ratios. We also make a conjecture on the classification of final motions of N-body problem, in which the superhyperbolic type of orbits is one of the building blocks., Comment: arXiv admin note: substantial text overlap with arXiv:2202.08534

Representing and synthesizing novel views in real-world dynamic scenes from casual monocular videos is a long-standing problem. Existing solutions typically approach dynamic scenes by applying geometry techniques or utilizing temporal information between several adjacent frames without considering the underlying background distribution in the entire scene or the transmittance over the ray dimension, limiting their performance on static and occlusion areas. Our approach $\textbf{D}$istribution-$\textbf{D}$riven neural radiance fields offers high-quality view synthesis and a 3D solution to $\textbf{D}$etach the background from the entire $\textbf{D}$ynamic scene, which is called $\text{D}^4$NeRF. Specifically, it employs a neural representation to capture the scene distribution in the static background and a 6D-input NeRF to represent dynamic objects, respectively. Each ray sample is given an additional occlusion weight to indicate the transmittance lying in the static and dynamic components. We evaluate $\text{D}^4$NeRF on public dynamic scenes and our urban driving scenes acquired from an autonomous-driving dataset. Extensive experiments demonstrate that our approach outperforms previous methods in rendering texture details and motion areas while also producing a clean static background. Our code will be released at https://github.com/Luciferbobo/D4NeRF.

In recent years, vision-centric perception has flourished in various autonomous driving tasks, including 3D detection, semantic map construction, motion forecasting, and depth estimation. Nevertheless, the latency of vision-centric approaches is too high for practical deployment (e.g., most camera-based 3D detectors have a runtime greater than 300ms). To bridge the gap between ideal research and real-world applications, it is necessary to quantify the trade-off between performance and efficiency. Traditionally, autonomous-driving perception benchmarks perform the offline evaluation, neglecting the inference time delay. To mitigate the problem, we propose the Autonomous-driving StreAming Perception (ASAP) benchmark, which is the first benchmark to evaluate the online performance of vision-centric perception in autonomous driving. On the basis of the 2Hz annotated nuScenes dataset, we first propose an annotation-extending pipeline to generate high-frame-rate labels for the 12Hz raw images. Referring to the practical deployment, the Streaming Perception Under constRained-computation (SPUR) evaluation protocol is further constructed, where the 12Hz inputs are utilized for streaming evaluation under the constraints of different computational resources. In the ASAP benchmark, comprehensive experiment results reveal that the model rank alters under different constraints, suggesting that the model latency and computation budget should be considered as design choices to optimize the practical deployment. To facilitate further research, we establish baselines for camera-based streaming 3D detection, which consistently enhance the streaming performance across various hardware. ASAP project page: https://github.com/JeffWang987/ASAP., Comment: code: https://github.com/JeffWang987/ASAP

We release a new codebase version of the BEVDet, dubbed branch dev2.0. With dev2.0, we propose BEVPoolv2 upgrade the view transformation process from the perspective of engineering optimization, making it free from a huge burden in both calculation and storage aspects. It achieves this by omitting the calculation and preprocessing of the large frustum feature. As a result, it can be processed within 0.82 ms even with a large input resolution of 640x1600, which is 15.1 times the previous fastest implementation. Besides, it is also less cache consumptive when compared with the previous implementation, naturally as it no longer needs to store the large frustum feature. Last but not least, this also makes the deployment to the other backend handy. We offer an example of deployment to the TensorRT backend in branch dev2.0 and show how fast the BEVDet paradigm can be processed on it. Other than BEVPoolv2, we also select and integrate some substantial progress that was proposed in the past year. As an example configuration, BEVDet4D-R50-Depth-CBGS scores 52.3 NDS on the NuScenes validation set and can be processed at a speed of 16.4 FPS with the PyTorch backend. The code has been released to facilitate the study on https://github.com/HuangJunJie2017/BEVDet/tree/dev2.0., Comment: Technique report

We study theoretically the zero temperature intrinsic anomalous Hall effect in an experimentally realized 2D spin-orbit coupled Bose gas. For anisotropic atomic interactions and as the spin-orbit coupling strength increases, the system undergoes a ground state phase transition from states exhibiting a total in-plane magnetization to those with a perpendicular magnetization along the $z$ direction. We show that finite frequency, or ac, Hall responses exist in both phases in the absence of an artificial magnetic field, as a result of finite inter-band transitions. However, the characteristics of the anomalous Hall responses are drastically different in these two phases because of the different symmetries preserved by the corresponding ground states. In particular, we find a finite dc Hall conductivity in one phase but not the other. The underlying physical reasons for this are analyzed further by exploring relations of the dc Hall conductivity to the system's chirality and Berry curvatures of the Bloch bands. Finally, we discuss an experimental method of probing the anomalous Hall effect in trapped systems., Comment: 13 pages, 11 figures

For bike sharing systems, demand prediction is crucial to ensure the timely re-balancing of available bikes according to predicted demand. Existing methods for bike sharing demand prediction are mostly based on its own historical demand variation, essentially regarding it as a closed system and neglecting the interaction between different transportation modes. This is particularly important for bike sharing because it is often used to complement travel through other modes (e.g., public transit). Despite some recent progress, no existing method is capable of leveraging spatiotemporal information from multiple modes and explicitly considers the distribution discrepancy between them, which can easily lead to negative transfer. To address these challenges, this study proposes a domain-adversarial multi-relational graph neural network (DA-MRGNN) for bike sharing demand prediction with multimodal historical data as input. A temporal adversarial adaptation network is introduced to extract shareable features from demand patterns of different modes. To capture correlations between spatial units across modes, we adapt a multi-relational graph neural network (MRGNN) considering both cross-mode similarity and difference. In addition, an explainable GNN technique is developed to understand how our proposed model makes predictions. Extensive experiments are conducted using real-world bike sharing, subway and ride-hailing data from New York City. The results demonstrate the superior performance of our proposed approach compared to existing methods and the effectiveness of different model components., Comment: arXiv admin note: substantial text overlap with arXiv:2203.10961

Understanding strongly correlated quantum materials, such as high $T_\textrm{c}$ superconductors, iron-based superconductors, and twisted bilayer graphene systems, remains to be one of the outstanding challenges in condensed matter physics. Quantum simulation with ultra-cold atoms in particular optical lattices, which provide orbital degrees of freedom, is a powerful tool to contribute new insights to this endeavor. Here, we report the experimental realization of an unconventional Bose-Einstein condensate of $^{87}$Rb atoms populating degenerate $p$-orbitals in a triangular optical lattice, exhibiting remarkably long coherence times. Using time-of-flight spectroscopy, we observe that this state spontaneously breaks the rotational symmetry and its momentum spectrum agrees with the theoretically predicted coexistence of exotic stripe and loop current orders. Like certain strongly correlated electronic systems with intertwined orders, as high-$T_\textrm{c}$ cuprate superconductors, twisted bilayer graphene, and the recently discovered chiral density-wave state in kagome superconductors $\textrm{AV}_3 \textrm{Sb}_5$ (A=K, Rb, Cs), the newly demonstrated quantum state, in spite of its markedly different energy scale and the bosonic quantum statistics, exhibits multiple symmetry breakings at ultralow temperatures. These findings hold the potential to enhance our comprehension of the fundamental physics governing these intricate quantum materials., Comment: 7+10 pages, 3+12 figures

Convolutional neural network (CNN) models have seen advanced improvements in performance in various domains, but lack of interpretability is a major barrier to assurance and regulation during operation for acceptance and deployment of AI-assisted applications. There have been many works on input interpretability focusing on analyzing the input-output relations, but the internal logic of models has not been clarified in the current mainstream interpretability methods. In this study, we propose a novel hybrid CNN-interpreter through: (1) An original forward propagation mechanism to examine the layer-specific prediction results for local interpretability. (2) A new global interpretability that indicates the feature correlation and filter importance effects. By combining the local and global interpretabilities, hybrid CNN-interpreter enables us to have a solid understanding and monitoring of model context during the whole learning process with detailed and consistent representations. Finally, the proposed interpretabilities have been demonstrated to adapt to various CNN-based model structures.

We train a deep learning artificial neural network model, Spatial Attention U-Net to recover useful ionospheric signals from noisy ionogram data measured by Hualien's Vertical Incidence Pulsed Ionospheric Radar. Our results show that the model can well identify F2 layer ordinary and extraordinary modes (F2o, F2x) and the combined signals of the E layer (ordinary and extraordinary modes and sporadic Es). The model is also capable of identifying some signals that were not labeled. The performance of the model can be significantly degraded by insufficient number of samples in the data set. From the recovered signals, we determine the critical frequencies of F2o and F2x and the intersection frequency between the two signals. The difference between the two critical frequencies is peaking at 0.63 MHz, with the uncertainty being 0.18 MHz., Comment: 17 pages, 7 figures, 3 tables

3D object detection with surrounding cameras has been a promising direction for autonomous driving. In this paper, we present SimMOD, a Simple baseline for Multi-camera Object Detection, to solve the problem. To incorporate multi-view information as well as build upon previous efforts on monocular 3D object detection, the framework is built on sample-wise object proposals and designed to work in a two-stage manner. First, we extract multi-scale features and generate the perspective object proposals on each monocular image. Second, the multi-view proposals are aggregated and then iteratively refined with multi-view and multi-scale visual features in the DETR3D-style. The refined proposals are end-to-end decoded into the detection results. To further boost the performance, we incorporate the auxiliary branches alongside the proposal generation to enhance the feature learning. Also, we design the methods of target filtering and teacher forcing to promote the consistency of two-stage training. We conduct extensive experiments on the 3D object detection benchmark of nuScenes to demonstrate the effectiveness of SimMOD and achieve new state-of-the-art performance. Code will be available at https://github.com/zhangyp15/SimMOD.

Self-supervised monocular methods can efficiently learn depth information of weakly textured surfaces or reflective objects. However, the depth accuracy is limited due to the inherent ambiguity in monocular geometric modeling. In contrast, multi-frame depth estimation methods improve the depth accuracy thanks to the success of Multi-View Stereo (MVS), which directly makes use of geometric constraints. Unfortunately, MVS often suffers from texture-less regions, non-Lambertian surfaces, and moving objects, especially in real-world video sequences without known camera motion and depth supervision. Therefore, we propose MOVEDepth, which exploits the MOnocular cues and VElocity guidance to improve multi-frame Depth learning. Unlike existing methods that enforce consistency between MVS depth and monocular depth, MOVEDepth boosts multi-frame depth learning by directly addressing the inherent problems of MVS. The key of our approach is to utilize monocular depth as a geometric priority to construct MVS cost volume, and adjust depth candidates of cost volume under the guidance of predicted camera velocity. We further fuse monocular depth and MVS depth by learning uncertainty in the cost volume, which results in a robust depth estimation against ambiguity in multi-view geometry. Extensive experiments show MOVEDepth achieves state-of-the-art performance: Compared with Monodepth2 and PackNet, our method relatively improves the depth accuracy by 20\% and 19.8\% on the KITTI benchmark. MOVEDepth also generalizes to the more challenging DDAD benchmark, relatively outperforming ManyDepth by 7.2\%. The code is available at https://github.com/JeffWang987/MOVEDepth., Comment: code: https://github.com/JeffWang987/MOVEDepth

Self-supervised monocular depth estimation is an attractive solution that does not require hard-to-source depth labels for training. Convolutional neural networks (CNNs) have recently achieved great success in this task. However, their limited receptive field constrains existing network architectures to reason only locally, dampening the effectiveness of the self-supervised paradigm. In the light of the recent successes achieved by Vision Transformers (ViTs), we propose MonoViT, a brand-new framework combining the global reasoning enabled by ViT models with the flexibility of self-supervised monocular depth estimation. By combining plain convolutions with Transformer blocks, our model can reason locally and globally, yielding depth prediction at a higher level of detail and accuracy, allowing MonoViT to achieve state-of-the-art performance on the established KITTI dataset. Moreover, MonoViT proves its superior generalization capacities on other datasets such as Make3D and DrivingStereo., Comment: Accepted by 3DV 2022

Deep supervision, or known as 'intermediate supervision' or 'auxiliary supervision', is to add supervision at hidden layers of a neural network. This technique has been increasingly applied in deep neural network learning systems for various computer vision applications recently. There is a consensus that deep supervision helps improve neural network performance by alleviating the gradient vanishing problem, as one of the many strengths of deep supervision. Besides, in different computer vision applications, deep supervision can be applied in different ways. How to make the most use of deep supervision to improve network performance in different applications has not been thoroughly investigated. In this paper, we provide a comprehensive in-depth review of deep supervision in both theories and applications. We propose a new classification of different deep supervision networks, and discuss advantages and limitations of current deep supervision networks in computer vision applications.

The anomalous Hall effect has had a profound influence on the understanding of many electronic topological materials but is much less studied in their bosonic counterparts. We predict that an intrinsic anomalous Hall effect exists in a recently realized bosonic chiral superfluid, a $p$-band Bose-Einstein condensate in a 2D hexagonal boron nitride optical lattice [X. Wang et al., https://www.nature.com/articles/s41586-021-03702-0 Nature (London) 596, 227 (2021)]. We evaluate the frequency-dependent Hall conductivity within a multi-orbital Bose-Hubbard model that accurately captures the real experimental system. We find that in the high frequency limit, the Hall conductivity is determined by finite loop current correlations on the $s$-orbital residing sublattice, the latter a defining feature of the system's chirality. In the opposite limit, the dc Hall conductivity can trace its origin back to the non-interacting band Berry curvature at the condensation momentum, although the contribution from atomic interactions can be significant. We discuss available experimental probes to observe this intrinsic anomalous Hall effect at both zero and finite frequencies., Comment: 6 pages, 4 figures

Coronal holes (CHs) are regions with unbalanced magnetic flux, and have been associated with open magnetic field (OMF) structures. However, it has been reported that some CHs do not intersect with OMF regions. To investigate the inconsistency, we apply a potential-field (PF) model to construct the magnetic fields of the coronal holes. As a comparison, we also use a thermodynamic magnetohydrodynamic (MHD) model to synthesize coronal images, and identify CHs from the synthetic images. The results from both the potential-field CHs and synthetic MHD CHs reveal that there is a significant percentage of closed field lines extending beyond the CH boundaries and more than 50% (17%) of PF (MHD) CHs do not contain OMF lines. The boundary-crossing field lines are more likely to be found in the lower latitudes during active times. While they tend to be located slightly closer than the non-boundary-crossing ones to the CH boundaries, nearly 40% (20%) of them in PF (MHD) CHs are not located in the boundary regions. The CHs without open field lines are often smaller and less unipolar than those with open field lines. The MHD model indicates higher temperature variations along the boundary-crossing field lines than the non-boundary crossing ones. The main difference between the results of the two models is that the dominant field lines in the PF and MHD CHs are closed and open field lines, respectively., Comment: accepted by ApJ