arXiv subjects

Xu Cao

Publications and source records attributed to Xu Cao.

At least 91 records · Page 5Linked to original sources

arXiv2023

SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration

We present SuperNormal, a fast, high-fidelity approach to multi-view 3D reconstruction using surface normal maps. With a few minutes, SuperNormal produces detailed surfaces on par with 3D scanners. We harness volume rendering to optimize a neural signed distance function (SDF) powered by multi-resolution hash encoding. To accelerate training, we propose directional finite difference and patch-based ray marching to approximate the SDF gradients numerically. While not compromising reconstruction quality, this strategy is nearly twice as efficient as analytical gradients and about three times faster than axis-aligned finite difference. Experiments on the benchmark dataset demonstrate the superiority of SuperNormal in efficiency and accuracy compared to existing multi-view photometric stereo methods. On our captured objects, SuperNormal produces more fine-grained geometry than recent neural 3D reconstruction methods.

cs.CV↗

arXiv2023

A Survey on Multimodal Large Language Models for Autonomous Driving

With the emergence of Large Language Models (LLMs) and Vision Foundation Models (VFMs), multimodal AI systems benefiting from large models have the potential to equally perceive the real world, make decisions, and control tools as humans. In recent months, LLMs have shown widespread attention in autonomous driving and map systems. Despite its immense potential, there is still a lack of a comprehensive understanding of key challenges, opportunities, and future endeavors to apply in LLM driving systems. In this paper, we present a systematic investigation in this field. We first introduce the background of Multimodal Large Language Models (MLLMs), the multimodal models development using LLMs, and the history of autonomous driving. Then, we overview existing MLLM tools for driving, transportation, and map systems together with existing datasets and benchmarks. Moreover, we summarized the works in The 1st WACV Workshop on Large Language and Vision Models for Autonomous Driving (LLVM-AD), which is the first workshop of its kind regarding LLMs in autonomous driving. To further promote the development of this field, we also discuss several important problems regarding using MLLMs in autonomous driving systems that need to be solved by both academia and industry.

cs.AI↗

arXiv2023

MACP: Efficient Model Adaptation for Cooperative Perception

Vehicle-to-vehicle (V2V) communications have greatly enhanced the perception capabilities of connected and automated vehicles (CAVs) by enabling information sharing to "see through the occlusions", resulting in significant performance improvements. However, developing and training complex multi-agent perception models from scratch can be expensive and unnecessary when existing single-agent models show remarkable generalization capabilities. In this paper, we propose a new framework termed MACP, which equips a single-agent pre-trained model with cooperation capabilities. We approach this objective by identifying the key challenges of shifting from single-agent to cooperative settings, adapting the model by freezing most of its parameters and adding a few lightweight modules. We demonstrate in our experiments that the proposed framework can effectively utilize cooperative observations and outperform other state-of-the-art approaches in both simulated and real-world cooperative perception benchmarks while requiring substantially fewer tunable parameters with reduced communication costs. Our source code is available at https://github.com/PurdueDigitalTwin/MACP.

cs.CV↗

arXiv2023

Receive, Reason, and React: Drive as You Say with Large Language Models in Autonomous Vehicles

The fusion of human-centric design and artificial intelligence (AI) capabilities has opened up new possibilities for next-generation autonomous vehicles that go beyond transportation. These vehicles can dynamically interact with passengers and adapt to their preferences. This paper proposes a novel framework that leverages Large Language Models (LLMs) to enhance the decision-making process in autonomous vehicles. By utilizing LLMs' linguistic and contextual understanding abilities with specialized tools, we aim to integrate the language and reasoning capabilities of LLMs into autonomous vehicles. Our research includes experiments in HighwayEnv, a collection of environments for autonomous driving and tactical decision-making tasks, to explore LLMs' interpretation, interaction, and reasoning in various scenarios. We also examine real-time personalization, demonstrating how LLMs can influence driving behaviors based on verbal commands. Our empirical results highlight the substantial advantages of utilizing chain-of-thought prompting, leading to improved driving decisions, and showing the potential for LLMs to enhance personalized driving experiences through ongoing verbal feedback. The proposed framework aims to transform autonomous vehicle operations, offering personalized support, transparent decision-making, and continuous learning to enhance safety and effectiveness. We achieve user-centric, transparent, and adaptive autonomous driving ecosystems supported by the integration of LLMs into autonomous vehicles.

cs.HC↗

arXiv2023

PIE: Simulating Disease Progression via Progressive Image Editing

Disease progression simulation is a crucial area of research that has significant implications for clinical diagnosis, prognosis, and treatment. One major challenge in this field is the lack of continuous medical imaging monitoring of individual patients over time. To address this issue, we develop a novel framework termed Progressive Image Editing (PIE) that enables controlled manipulation of disease-related image features, facilitating precise and realistic disease progression simulation. Specifically, we leverage recent advancements in text-to-image generative models to simulate disease progression accurately and personalize it for each patient. We theoretically analyze the iterative refining process in our framework as a gradient descent with an exponentially decayed learning rate. To validate our framework, we conduct experiments in three medical imaging domains. Our results demonstrate the superiority of PIE over existing methods such as Stable Diffusion Walk and Style-Based Manifold Extrapolation based on CLIP score (Realism) and Disease Classification Confidence (Alignment). Our user study collected feedback from 35 veteran physicians to assess the generated progressions. Remarkably, 76.2% of the feedback agrees with the fidelity of the generated progressions. To our best knowledge, PIE is the first of its kind to generate disease progression images meeting real-world standards. It is a promising tool for medical research and clinical practice, potentially allowing healthcare providers to model disease trajectories over time, predict future treatment responses, and improve patient outcomes.

eess.IV↗

arXiv2023

Data driven isospin analysis of timelike octet baryons electromagnetic form factors and charmonium decay into baryon-anti-baryon

Inspired by the recent precise data, we perform model-independently an isospin decomposition of the timelike octet baryons electromagnetic form factors. As noted in our previous work, the relative magnitude of isoscalar and isovector component is determined with the input of data on various isospin channels. Herein we further assert that {their relative phase can be constraint by the phase difference of oscillatory modulation of effective form factors} between isospin channels. The framework is extended to analyze the data of differential cross sections and applied to the form factors of nucleon and hyperons with detail and isospin non-conservation of charmonium decay into baryon-anti-baryon as well. We address that isospin analysis is meaningful when the isospin broken scale is compared to or smaller than the uncertainties of data.

hep-ph↗

arXiv2023

Drive as You Speak: Enabling Human-Like Interaction with Large Language Models in Autonomous Vehicles

The future of autonomous vehicles lies in the convergence of human-centric design and advanced AI capabilities. Autonomous vehicles of the future will not only transport passengers but also interact and adapt to their desires, making the journey comfortable, efficient, and pleasant. In this paper, we present a novel framework that leverages Large Language Models (LLMs) to enhance autonomous vehicles' decision-making processes. By integrating LLMs' natural language capabilities and contextual understanding, specialized tools usage, synergizing reasoning, and acting with various modules on autonomous vehicles, this framework aims to seamlessly integrate the advanced language and reasoning capabilities of LLMs into autonomous vehicles. The proposed framework holds the potential to revolutionize the way autonomous vehicles operate, offering personalized assistance, continuous learning, and transparent decision-making, ultimately contributing to safer and more efficient autonomous driving technologies.

cs.HC↗

arXiv2023

Mitigating Transformer Overconfidence via Lipschitz Regularization

Though Transformers have achieved promising results in many computer vision tasks, they tend to be over-confident in predictions, as the standard Dot Product Self-Attention (DPSA) can barely preserve distance for the unbounded input domain. In this work, we fill this gap by proposing a novel Lipschitz Regularized Transformer (LRFormer). Specifically, we present a new similarity function with the distance within Banach Space to ensure the Lipschitzness and also regularize the term by a contractive Lipschitz Bound. The proposed method is analyzed with a theoretical guarantee, providing a rigorous basis for its effectiveness and reliability. Extensive experiments conducted on standard vision benchmarks demonstrate that our method outperforms the state-of-the-art single forward pass approaches in prediction, calibration, and uncertainty estimation.

cs.LG↗

arXiv2023

Accessing Compton Form Factors at the Electron Ion Collider in China: An Impact study on $\textbf{Im} {\mathcal E}$

We estimate the impact of asymmetry measurements of Deeply Virtual Compton Scattering (DVCS) with transversely polarized proton beam taken at a future Electron Ion Collider in China (EicC) on the extraction of Compton Form Factors (CFFs). The CFFs extracted from an analysis based on artificial neural-network approach are reweighted by means of pseudo-data generated in the expected kinematic region of EicC. We find a remarkable improvement in the extraction of CFF $\textbf{Im} {\mathcal E}$, especially at the range of parton momentum fraction $x \sim$ 0.01, thus hinting for a future experimental probe of the parton orbital angular momentum. This work casts a glance at a practical implementation of Bayesian reweighting method on CFFs' impact study.

hep-ph↗

arXiv2023

CEMFormer: Learning to Predict Driver Intentions from In-Cabin and External Cameras via Spatial-Temporal Transformers

Driver intention prediction seeks to anticipate drivers' actions by analyzing their behaviors with respect to surrounding traffic environments. Existing approaches primarily focus on late-fusion techniques, and neglect the importance of maintaining consistency between predictions and prevailing driving contexts. In this paper, we introduce a new framework called Cross-View Episodic Memory Transformer (CEMFormer), which employs spatio-temporal transformers to learn unified memory representations for an improved driver intention prediction. Specifically, we develop a spatial-temporal encoder to integrate information from both in-cabin and external camera views, along with episodic memory representations to continuously fuse historical data. Furthermore, we propose a novel context-consistency loss that incorporates driving context as an auxiliary supervision signal to improve prediction performance. Comprehensive experiments on the Brain4Cars dataset demonstrate that CEMFormer consistently outperforms existing state-of-the-art methods in driver intention prediction.

cs.CV↗

arXiv2023

Toy model to understand oscillatory behavior in timelike nucleon form factors

To understand the oscillatory behavior exhibited in the timelike electromagnetic form factors of nucleons, we propose a toy model based on the Jost function of the $N\bar N$ pair into the timelike form factors with the help of the distorted-wave Born approximation. By constructing a simple square-well potential reflecting the final-state interaction of $N\bar N$, we naturally represent the damped oscillatory phenomenon in the timelike electromagnetic form factors of nucleons. Especially, our study reveals that the ``period" of the oscillation is approximately determined by the Yukawa interaction range $1/m_π$. Other possible potentials are also discussed. The threshold enhancements of the cross sections for $e^+e^-\to n\bar n$, $Λ\barΛ$, and $Λ_c\barΛ_c$ can also be understand within this scenario.

nucl-th↗

arXiv2023

Multi-View Azimuth Stereo via Tangent Space Consistency

We present a method for 3D reconstruction only using calibrated multi-view surface azimuth maps. Our method, multi-view azimuth stereo, is effective for textureless or specular surfaces, which are difficult for conventional multi-view stereo methods. We introduce the concept of tangent space consistency: Multi-view azimuth observations of a surface point should be lifted to the same tangent space. Leveraging this consistency, we recover the shape by optimizing a neural implicit surface representation. Our method harnesses the robust azimuth estimation capabilities of photometric stereo methods or polarization imaging while bypassing potentially complex zenith angle estimation. Experiments using azimuth maps from various sources validate the accurate shape recovery with our method, even without zenith angles.

cs.CV↗

arXiv2023

ECON: Explicit Clothed humans Optimized via Normal integration

The combination of deep learning, artist-curated scans, and Implicit Functions (IF), is enabling the creation of detailed, clothed, 3D humans from images. However, existing methods are far from perfect. IF-based methods recover free-form geometry, but produce disembodied limbs or degenerate shapes for novel poses or clothes. To increase robustness for these cases, existing work uses an explicit parametric body model to constrain surface reconstruction, but this limits the recovery of free-form surfaces such as loose clothing that deviates from the body. What we want is a method that combines the best properties of implicit representation and explicit body regularization. To this end, we make two key observations: (1) current networks are better at inferring detailed 2D maps than full-3D surfaces, and (2) a parametric model can be seen as a "canvas" for stitching together detailed surface patches. Based on these, our method, ECON, has three main steps: (1) It infers detailed 2D normal maps for the front and back side of a clothed person. (2) From these, it recovers 2.5D front and back surfaces, called d-BiNI, that are equally detailed, yet incomplete, and registers these w.r.t. each other with the help of a SMPL-X body mesh recovered from the image. (3) It "inpaints" the missing geometry between d-BiNI surfaces. If the face and hands are noisy, they can optionally be replaced with the ones of SMPL-X. As a result, ECON infers high-fidelity 3D humans even in loose clothes and challenging poses. This goes beyond previous methods, according to the quantitative evaluation on the CAPE and Renderpeople datasets. Perceptual studies also show that ECON's perceived realism is better by a large margin. Code and models are available for research purposes at econ.is.tue.mpg.de

cs.CV↗

arXiv2023

ViTASD: Robust Vision Transformer Baselines for Autism Spectrum Disorder Facial Diagnosis

Autism spectrum disorder (ASD) is a lifelong neurodevelopmental disorder with very high prevalence around the world. Research progress in the field of ASD facial analysis in pediatric patients has been hindered due to a lack of well-established baselines. In this paper, we propose the use of the Vision Transformer (ViT) for the computational analysis of pediatric ASD. The presented model, known as ViTASD, distills knowledge from large facial expression datasets and offers model structure transferability. Specifically, ViTASD employs a vanilla ViT to extract features from patients' face images and adopts a lightweight decoder with a Gaussian Process layer to enhance the robustness for ASD analysis. Extensive experiments conducted on standard ASD facial analysis benchmarks show that our method outperforms all of the representative approaches in ASD facial analysis, while the ViTASD-L achieves a new state-of-the-art. Our code and pretrained models are available at https://github.com/IrohXu/ViTASD.

cs.CV↗

arXiv2023

Photo- and electro-production of narrow exotic states: from light quarks to charm and up to bottom

Accessing a full image of the inner content of hadrons represents a central endeavour of modern particle physics, with the main scientific motivation to investigate the strong interaction binding the visible matter. On the one hand, the structure of known exotic candidates is a fundamental open issue addressed widely by scientists. On the other hand, looking for new states of exotic nature is a central component for theoretical and experimental efforts from electron-positron machine and electron accelerator with fixed target to heavy ion and electron-ion colliders. In this article we present a succinct short overview of the attempt to search for exotic narrow $N^*$ and $Z$ states containing light quarks only or also charm, and its connotation for bottom regions (the latter two are also known as $P_c$ ($Z_c$) and $P_b$ ($Z_b$) states, respectively in {the literature}). We address the effort of searching for exotic narrow $N^*$ and $Z$ states in light quark sector. We focus on recent progress in searching for signal of $P_c$ and $Z_c$ states photoproduction and its implication into the $P_b$ and $Z_b$ photoproduction and their decay properties. We also discuss future perspectives for the field in electron-ion colliders, a good place to disentangle the nature of some of these states and investigate some other enlightening topics including QCD trace anomaly and quarkonium-nucleon scattering length.

hep-ph↗

arXiv2023

SuperScaler: Supporting Flexible DNN Parallelization via a Unified Abstraction

With the growing model size, deep neural networks (DNN) are increasingly trained over massive GPU accelerators, which demands a proper parallelization plan that transforms a DNN model into fine-grained tasks and then schedules them to GPUs for execution. Due to the large search space, the contemporary parallelization plan generators often rely on empirical rules that couple transformation and scheduling, and fall short in exploring more flexible schedules that yield better memory usage and compute efficiency. This tension can be exacerbated by the emerging models with increasing complexity in their structure and model size. SuperScaler is a system that facilitates the design and generation of highly flexible parallelization plans. It formulates the plan design and generation into three sequential phases explicitly: model transformation, space-time scheduling, and data dependency preserving. Such a principled approach decouples multiple seemingly intertwined factors and enables the composition of highly flexible parallelization plans. As a result, SuperScaler can not only generate empirical parallelization plans, but also construct new plans that achieve up to 3.5X speedup compared to state-of-the-art solutions like DeepSpeed, Megatron and Alpa, for emerging DNN models like Swin-Transformer and AlphaFold2, as well as well-optimized models like GPT-3.

cs.DC↗

arXiv2022

Comparing fast imaging techniques for individual pulse imaging by Cherenkov in vivo from electron FLASH irradiation

Objective: In this study, a fast imaging technique was developed for the first in vivo Cherenkov emission imaging from an ultra-high dose rate (UHDR) electron beam source at single pulse (360 Hz) submillimeter resolution. Approach: A CMOS camera, gated to the UHDR LINAC, imaged the Cherenkov emission profiles pulse by pulse passively during the irradiation of mice on their limbs and intestinal region. The utility of an intensifier was investigated for its effect on image quality including signal to noise and spatial resolution. Pulse by pulse variability in Cherenkov emission profile were quantified spatially and temporally. Main results: An intensifier improved the emission profile signal to noise ratio from 15 to 280, with reduced spatial resolution. The profile extended beyond of the treatment field due to the lateral scattering of the electrons in tissue and its optical properties. The CMOS camera with an intensifier detected the changes in Cherenkov emission profile during expiration and inspiration of the respiration cycle for the mice to be about 3 mm. Significance: This fast imaging technique can be utilized for in vivo intrafraction monitoring of FLASH patient treatments at single pulse resolution. It can display delivery differences during respiration, and variability in the delivered treatment's surface profile, which may perturb from the intended UHDR treatment more for pencil beam scanning systems. The technique may leverage Cherenkov emission surface profile to gate the treatment delivery via respiratory gating systems under FLASH conditions.

physics.med-ph↗

arXiv2022

THMA: Tencent HD Map AI System for Creating HD Map Annotations

Nowadays, autonomous vehicle technology is becoming more and more mature. Critical to progress and safety, high-definition (HD) maps, a type of centimeter-level map collected using a laser sensor, provide accurate descriptions of the surrounding environment. The key challenge of HD map production is efficient, high-quality collection and annotation of large-volume datasets. Due to the demand for high quality, HD map production requires significant manual human effort to create annotations, a very time-consuming and costly process for the map industry. In order to reduce manual annotation burdens, many artificial intelligence (AI) algorithms have been developed to pre-label the HD maps. However, there still exists a large gap between AI algorithms and the traditional manual HD map production pipelines in accuracy and robustness. Furthermore, it is also very resource-costly to build large-scale annotated datasets and advanced machine learning algorithms for AI-based HD map automatic labeling systems. In this paper, we introduce the Tencent HD Map AI (THMA) system, an innovative end-to-end, AI-based, active learning HD map labeling system capable of producing and labeling HD maps with a scale of hundreds of thousands of kilometers. In THMA, we train AI models directly from massive HD map datasets via supervised, self-supervised, and weakly supervised learning to achieve high accuracy and efficiency required by downstream users. THMA has been deployed by the Tencent Map team to provide services to downstream companies and users, serving over 1,000 labeling workers and producing more than 30,000 kilometers of HD map data per day at most. More than 90 percent of the HD map data in Tencent Map is labeled automatically by THMA, accelerating the traditional HD map labeling process by more than ten times.

cs.CV↗