About

Shenlong Wang is an Assistant Professor of Computer Science at the University of Illinois Urbana-Champaign. His research focuses on computer vision and robotics, with an emphasis on understanding, modeling, and simulating the physical world for content creation, robotics, and AI for science. His work has been recognized with awards and finalist recognitions at conferences such as ISMAR, IROS, and CVPR, and he is a recipient of the NSF CAREER Award. Wang's contributions include advancing the understanding of autonomous systems and simulation, contributing to the development of perception and modeling techniques in robotics and computer vision.

Research topics

• Computer Science
• Artificial Intelligence
• Computer vision
• Geography
• Engineering
• Remote sensing
• Algorithm
• Database
• Computer graphics (images)

Selected publications

• Tri-axial low-frequency vibration isolator with decoupled dynamics

  International Journal of Mechanical Sciences · 2026-03-31 · 1 citations

  articleCorresponding
  PublisherDOI

• Quasi-zero-stiffness vibration isolation: A comprehensive review of design principles and engineering applications

  International Journal of Non-Linear Mechanics · 2026-05-19

  articleOpen accessSenior authorCorresponding

  Quasi-zero-stiffness (QZS) vibration isolation technology achieves high-static-low-dynamic-stiffness (HSLDS) and thereby resolves the fundamental trade-off between load-bearing capacity and isolation frequency inherent to conventional linear isolators. This paper systematically reviews recent advances in QZS isolators with a focus on literature published in the last five years. Systems are classified by degrees of freedom (DOF) into single-DOF (SDOF) and multi-DOF (MDOF) configurations. For SDOF systems, seven principal design archetypes are discussed: multi-spring, spring-linkage, cam-roller, magnetic, biomimetic, origami-inspired, and metamaterial-based structures. Beyond the classic approach of assembling positive-stiffness and negative-stiffness components, emerging design strategies such as topology optimisation and meta-structure concepts are examined. Adjustable and load-adaptive designs that preserve stable high-performance isolation under variable static loading receive dedicated emphasis. As QZS isolators are inherently nonlinear systems, their complex dynamic behaviours including sub-harmonic and super-harmonic resonances, bifurcations and chaos are analysed. For MDOF systems, progress in 2-DOF, 3-DOF and 6-DOF platforms is summarised with attention to multi-dimensional vibration decoupling and coordinated stiffness regulation for high-precision equipment. Finally, engineering applications of QZS technology are highlighted covering vehicle seat isolation, aerospace systems, vibration energy harvesting, bridge seismic protection and precision manufacturing, demonstrating substantial practical utility. • Comprehensive review of quasi-zero-stiffness vibration isolators. • Covers SDOF designs: multi-spring, linkage, cam, magnetic, bionic, origami, metamaterial. • Summarizes MDOF isolators for 2-, 3-, 6-DOF vibration control. • Highlights adjustable and load-adaptive QZS for variable loads. • Discusses engineering applications in seats, aerospace, energy harvesting, bridges.

  PublisherDOI

• Dual Structure Reinforces Interfacial Polarized MXene/PVDF-TrFE Piezoelectric Nanocomposite for Pressure Monitoring

  Nano-Micro Letters · 2025-07-03 · 29 citations

  articleOpen access

  The emerging interfacial polarization strategy exhibits applicative potential in piezoelectric enhancement. However, there is an ongoing effort to address the inherent limitations arising from charge bridging phenomena and stochastic interface disorder that plague the improvement of piezoelectric performance. Here, we report a dual structure reinforced MXene/PVDF-TrFE piezoelectric composite, whose piezoelectricity is enhanced under the coupling effect of interfacial polarization and structural design. Synergistically, molecular dynamics simulations, density functional theory calculations and experimental validation revealed the details of interfacial interactions, which promotes the net spontaneous polarization of PVDF-TrFE from the 0.56 to 31.41 Debye. The oriented MXene distribution and porous structure not only tripled the piezoelectric response but also achieved an eightfold increase in sensitivity within the low-pressure region, along with demonstrating cyclic stability exceeding 20,000 cycles. The properties reinforcement originating from dual structure is elucidated through the finite element simulation and experimental validation. Attributed to the excellent piezoelectric response and deep learning algorithm, the sensor can effectively recognize the signals of artery pulse and finger flexion. Finally, a 3 × 3 sensor array is fabricated to monitor the pressure distribution wirelessly. This study provides an innovative methodology for reinforcing interfacial polarized piezoelectric materials and insight into structural designs.

  PublisherOA PDFDOI

• Plenoptic PNG: Real-Time Neural Radiance Fields in 150 KB

  2025-03-25

  articleSenior author

  The goal of this paper is to encode a 3D scene into an extremely compact representation from <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$2 D$</tex> images and to enable its transmittance, decoding and rendering in real-time across various platforms. Despite the progress in NeRFs and Gaussian Splats, their large model size and specialized renderers make it challenging to distribute free-viewpoint 3D content as easily as images. To address this, we have designed a novel 3D representation that encodes the plenoptic function into sinusoidal function indexed dense volumes. This approach facilitates feature sharing across different locations, improving compactness over traditional spatial voxels. The memory footprint of the dense 3D feature grid can be further reduced using spatial decomposition techniques. This design combines the strengths of spatial hashing functions and voxel decomposition, resulting in a model size as small as 150 KB for each 3D scene. Moreover, PPNG features a lightweight rendering pipeline with only 300 lines of code that decodes its representation into standard GL textures and fragment shaders. This enables realtime rendering using the traditional GL pipeline, ensuring universal compatibility and efficiency across various platforms without additional dependencies. Our results are available at: https://jyl.kr/ppng

  PublisherDOI

• LidarDM: Generative LiDAR Simulation in a Generated World

  2025-05-19 · 4 citations

  articleSenior author

  We present LidarDM, a novel LiDAR generative model capable of producing realistic, layout-aware, physically plausible, and temporally coherent LiDAR videos. LidarDM stands out with two unprecedented capabilities in LiDAR generative modeling: (i) LiDAR generation guided by driving scenarios, offering significant potential for autonomous driving simulations, and (ii) 4D LiDAR point cloud generation, enabling the creation of realistic and temporally coherent sequences. At the heart of our model is a novel integrated 4D world generation framework. Specifically, we employ latent diffusion models to generate the 3D scene, combine it with dynamic actors to form the underlying 4D world, and subsequently produce realistic sensory observations within this virtual environment. Our experiments indicate that our approach outperforms competing algorithms in realism, temporal coherency, and layout consistency. We additionally show that LidarDM can be used as a generative world model simulator for training and testing perception models. We release our source code and checkpoints at https://github.com/vzyrianov/LidarDM

  PublisherDOI

• UrbanIR: Large-Scale Urban Scene Inverse Rendering from a Single Video

  2025-03-25 · 1 citations

  articleSenior author

  We present UrbanIR (Urban Scene Inverse Rendering), a new inverse graphics model that enables realistic, free-viewpoint renderings of scenes under various lighting conditions with a single video. It accurately infers shape, albedo, visibility, and sun and sky illumination from wide-baseline videos, such as those from car-mounted cameras, differing from NeRF's dense view settings. In this context, standard methods often yield subpar geometry and material estimates, such as inaccurate roof representations and numerous ‘floaters’. UrbanIR addresses these issues with novel losses that reduce errors in inverse graphics inference and rendering artifacts. Its techniques allow for precise shadow volume estimation in the original scene. The model's outputs support controllable editing, enabling photorealistic free-viewpoint renderings of night simulations, relit scenes, and inserted objects, marking a significant improvement over existing state-of-the-art methods. Our code and data will be made publicly available upon acceptance.

  PublisherDOI

• MonoPatchNeRF: Improving Neural Radiance Fields with Patch-Based Monocular Guidance

  2025-03-25 · 1 citations

  article

  The latest regularized Neural Radiance Field (NeRF) approaches produce poor geometry and view extrapolation for large scale sparse view scenes, such as ETH3D. Density-based approaches tend to be under-constrained, while surface-based approaches tend to miss details. In this paper, we take a density-based approach, sampling patches instead of individual rays to better incorporate monocular depth and normal estimates and patch-based photometric consistency constraints between training views and sampled virtual views. Loosely constraining densities based on estimated depth aligned to sparse points further improves geometric accuracy. While maintaining similar view synthesis quality, our approach significantly improves geometric accuracy on the ETH3D benchmark, e.g. increasing the F1@2cm score by 4x-8x compared to other regularized density-based approaches, with much lower training and inference time than other approaches.

  PublisherDOI

• IRIS: Inverse Rendering of Indoor Scenes from Low Dynamic Range Images

  2025-06-10 · 2 citations

  article

  Inverse rendering seeks to recover 3D geometry, surface material, and lighting from captured images, enabling advanced applications such as novel-view synthesis, relighting, and virtual object insertion. However, most existing techniques rely on high dynamic range (HDR) images as input, limiting accessibility for general users. In response, we introduce IRIS, an inverse rendering framework that recovers the physically based material, spatially-varying HDR lighting, and camera response functions from multi-view, low-dynamic-range (LDR) images. By eliminating the dependence on HDR input, we make inverse rendering technology more accessible. We evaluate our approach on real-world and synthetic scenes and compare it with state-of-the-art methods. Our results show that IRIS effectively recovers HDR lighting, accurate material, and plausible camera response functions, supporting photorealistic relighting and object insertion.

  PublisherDOI

• Plasticized electrohydraulic robot autopilots in the deep sea

  Science Robotics · 2025-08-13 · 14 citations

  article

  Soft robots, with their compliant bodies, minimal environmental disturbance, and ability to withstand ambient pressures, offer promising solutions for deep-sea exploration. However, a common challenge of stiffening in soft materials impairs their effective actuation in harsh conditions. In this work, we integrated a liquid dielectric plasticizer within an electrohydraulic soft robot, serving dual critical functions as a softening agent to maintain the softness of the polymer shell and an electrohydraulic fluid for efficient actuation. In addition, by using the surrounding seawater as alternating electrodes, we prevented charge retention in dielectric layers, enabling sustained actuation performance. Field tests at depths of ~1360, 3176, and ~4071 meters confirmed the robot's ability to sense the environment, navigate complex trajectories, and withstand unsteady disturbances. Our work offers a generalized and straightforward framework for developing soft materials tailored for deep-sea applications, paving the way for soft robots to execute real-world missions.

  PublisherDOI

• Map Space Belief Prediction for Manipulation-Enhanced Mapping

  ArXiv.org · 2025-02-28

  preprintOpen access

  Searching for objects in cluttered environments requires selecting efficient viewpoints and manipulation actions to remove occlusions and reduce uncertainty in object locations, shapes, and categories. In this work, we address the problem of manipulation-enhanced semantic mapping, where a robot has to efficiently identify all objects in a cluttered shelf. Although Partially Observable Markov Decision Processes~(POMDPs) are standard for decision-making under uncertainty, representing unstructured interactive worlds remains challenging in this formalism. To tackle this, we define a POMDP whose belief is summarized by a metric-semantic grid map and propose a novel framework that uses neural networks to perform map-space belief updates to reason efficiently and simultaneously about object geometries, locations, categories, occlusions, and manipulation physics. Further, to enable accurate information gain analysis, the learned belief updates should maintain calibrated estimates of uncertainty. Therefore, we propose Calibrated Neural-Accelerated Belief Updates (CNABUs) to learn a belief propagation model that generalizes to novel scenarios and provides confidence-calibrated predictions for unknown areas. Our experiments show that our novel POMDP planner improves map completeness and accuracy over existing methods in challenging simulations and successfully transfers to real-world cluttered shelves in zero-shot fashion.

  PublisherOA PDFDOI

Frequent coauthors

• Raquel Urtasun

  90 shared

• Wei-Chiu Ma

  33 shared

• Gellért Máttyus

  16 shared

• Kelvin Wong

  Houston Methodist

  15 shared

• Sivabalan Manivasagam

  14 shared

• Ioan Andrei Bârsan

  14 shared

• Sanja Fidler

  12 shared

• Min Bai

  12 shared

Labs

• Siebel School of Computing and Data SciencePI

Education

• Ph.D./Dr., School of Mechanical Engineering

  University of Shanghai for Science and Technology

Awards & honors

• Teacher Ranked as Excellent, Computer Science, University of…
• ISMAR Best Paper (2025)
• Dean's Award for Excellence in Research, Grainger College of…
• Faculty Early Career Development Program (CAREER), National…
• Amazon Faculty Research Award (2022)