About

Bedrich Benes is a Professor of Computer Science at Purdue University and serves as Associate Department Head. He received his Ph.D. in Computer Science from Czech Technical University in Prague in 1998, and also holds an MS from the same institution. His research focuses on generative methods for geometry synthesis, including procedural and inverse procedural modeling, the simulation of natural phenomena, and additive manufacturing. Dr. Benes has published over 200 research papers in these fields and has been sponsored by organizations such as the National Science Foundation, NASA, Adobe Research, Intel, Siemens, Samsung, the Department of Energy, and Ford Inc. He has held significant roles in the academic community, including co-chairing Eurographics 2017 and participating as a Program Committee Member for numerous conferences such as Siggraph, Siggraph Asia, and the International Conference on Geometric Modeling and Processing. Dr. Benes is a senior member of ACM and IEEE, a Eurographics Fellow, and serves as the editor-in-chief of Elsevier's Graphical Models. His editorial work also includes positions on various journal editorial boards. His contributions to the field include work on procedural modeling, natural phenomena simulation, and additive manufacturing, making him a prominent figure in computer graphics and visualization.

Research topics

• Computer science
• Artificial intelligence
• Computer graphics (images)
• Computer vision
• Human–computer interaction

Selected publications

• VR BioTalk—Hands-Free Visual Analytics of Phenotyping Data Using Natural Conversation

  IEEE Access · 2026-01-01

  articleOpen accessSenior author

  We introduce, implement, and test <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VR BioTalk</i>, a hands-free, immersive, voice-controlled visual analytics system for phenotypic data. Our system does not require any programming knowledge. Yet, it enables users to receive an interactive solution to complex tasks involving large datasets through simple verbal commands, such as ‘‘Show me all leaves smaller than the average and calculate their leaf area index.’’ We claim three main contributions: (1) preprocessing and feature extraction of point cloud data for interactive visual analytics, (2) development of a novel interface that converts user speech into commands, and (3) an immersive VR visualization that executes the commands and displays the results in VR. The speech recognition system’s precision has been validated on 416 spoken commands across 13 English accents, with an accuracy of around 99.7% for transcription and 94% for command recognition. The visualization averages 63 FPS, and the system’s response time is approximately 1.25 seconds. We tested <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VR BioTalk</i> on several tasks that would otherwise require extensive programming knowledge. We tested our system with 9 participants, and the results show that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VR BioTalk</i> is highly usable, engaging, and easy to use, enabling experts with no programming background to explore large phenotyping datasets and generate hypotheses in natural language.

  PublisherDOI

• Neural Assistive Impulses: Synthesizing Exaggerated Motions for Physics-based Characters

  arXiv (Cornell University) · 2026-04-07

  preprintOpen accessSenior author

  Physics-based character animation has become a fundamental approach for synthesizing realistic, physically plausible motions. While current data-driven deep reinforcement learning (DRL) methods can synthesize complex skills, they struggle to reproduce exaggerated, stylized motions, such as instantaneous dashes or mid-air trajectory changes, which are required in animation but violate standard physical laws. The primary limitation stems from modeling the character as an underactuated floating-base system, in which internal joint torques and momentum conservation strictly govern motion. Direct attempts to enforce such motions via external wrenches often lead to training instability, as velocity discontinuities produce sparse, high-magnitude force spikes that prevent policy convergence. We propose Assistive Impulse Neural Control, a framework that reformulates external assistance in impulse space rather than force space to ensure numerical stability. We decompose the assistive signal into an analytic high-frequency component derived from Inverse Dynamics and a learned low-frequency residual correction, governed by a hybrid neural policy. We demonstrate that our method enables robust tracking of highly agile, dynamically infeasible maneuvers that were previously intractable for physics-based methods.

  PublisherDOI

• Spatial Exploration Behavior in XR Learning: Toward Passive Assessment of Embodied Engagement

  2026-03-21

  article
  PublisherDOI

• Loops2Roofs: Diffusion-based 3D Roof Generation using a Loop Representation

  ACM Transactions on Graphics · 2026-04-10

  articleOpen access

  Effectively representing and generating 3D roof structures remains a challenging task in urban modeling. This paper introduces a loop-based roof representation (LFR) and Loops2Roofs algorithm, a novel diffusion-based generative approach that directly outputs 3D polygonal roof meshes using the LFR. Our key novelty lies in representing the roofs as a set of polygonal faces bounded by loops. This flexible representation improves generalization to complex roof structures across different architectural styles and can be effectively learned and generated by machine learning models. Roof generation proceeds in two stages. First, a Transformer-based diffusion model directly denoises the 3D coordinates of roof face vertices. This model is conditioned on structural priors, specifically the number of roof faces and vertices per face, which are pre-generated using an auto-regressive model. Second, a neural stitching module enforces roof topology and recovers the geometric incident relationships between edges of different 3D loops. Optionally, our algorithm can incorporate 2D building footprints as input for image-conditioned roof generation. The final output is a compact, structured roof mesh encoded in the LFR format. We demonstrate the effectiveness and efficiency of our approach by generating diverse, realistic roof models, ranging from synthetic to real-world buildings. Experiments show that Loops2Roofs significantly advances structured roof generation, outperforming existing methods on three benchmarks.

  PublisherDOI

• Neural Assistive Impulses: Synthesizing Exaggerated Motions for Physics-based Characters

  arXiv (Cornell University) · 2026-04-07

  articleOpen accessSenior author

  Physics-based character animation has become a fundamental approach for synthesizing realistic, physically plausible motions. While current data-driven deep reinforcement learning (DRL) methods can synthesize complex skills, they struggle to reproduce exaggerated, stylized motions, such as instantaneous dashes or mid-air trajectory changes, which are required in animation but violate standard physical laws. The primary limitation stems from modeling the character as an underactuated floating-base system, in which internal joint torques and momentum conservation strictly govern motion. Direct attempts to enforce such motions via external wrenches often lead to training instability, as velocity discontinuities produce sparse, high-magnitude force spikes that prevent policy convergence. We propose Assistive Impulse Neural Control, a framework that reformulates external assistance in impulse space rather than force space to ensure numerical stability. We decompose the assistive signal into an analytic high-frequency component derived from Inverse Dynamics and a learned low-frequency residual correction, governed by a hybrid neural policy. We demonstrate that our method enables robust tracking of highly agile, dynamically infeasible maneuvers that were previously intractable for physics-based methods.

  PublisherOA PDF

• Where Are the City Trees? Monitoring Urban Trees across the U.S. Using Generative AI

  Communications of the ACM · 2026-03-30

  articleOpen access

  Finding where the trees are in a city and monitoring any changes are essential for sustainable urban management. Historically, urban forests are mainly inventoried via manual processes often limited to public lands. Leveraging advances in computing, we present a novel generative artificial intelligence (AI) method along with a first-ever national-scale dataset, to automatically localize trees in cities across the nation using satellite imagery. Our monitoring approach is fully automated and can be completed for 330 U.S. cities within less than a day of computing, enabling actionable knowledge of changes in urban trees and supporting sustainable development decisions. We successfully localized and counted over 278 million trees, achieving an average tree count accuracy of 92.5% and spatial accuracy of 1.5m for 2024–2025. Our computational approach allows for novel nationwide analysis to be performed. For example, we can localize approximately 117 million trees on private lands and 161 million on public lands. Further, we show and quantify that urban tree distribution exhibits strong spatial disparity, with low-income communities having substantially fewer trees and less canopy cover than others. In addition, we compare tree count and layouts before and after multiple major events (e.g., major fires and destructive weather phenomena). Overall, our approach enhances computational urban planning, including weather and extreme event forecasting, for the development of sustainable cities.

  PublisherDOI

• LatentMimic: Terrain-Adaptive Locomotion via Latent Space Imitation

  ArXiv.org · 2026-04-07

  articleOpen accessSenior author

  Developing natural and diverse locomotion controllers for quadruped robots that can adapt to complex terrains while preserving motion style remains a significant challenge. Existing imitation-based methods face a fundamental optimization trade-off: strict adherence to motion capture (mocap) references penalizes the geometric deviations required for terrain adaptability, whereas terrain-centric policies often compromise stylistic fidelity. We introduce LatentMimic, a novel locomotion learning framework that decouples stylistic fidelity from geometric constraints. By minimizing the marginal latent divergence between the policy's state-action distribution and a learned mocap prior, our approach provides a conditional relaxation of rigid pose-tracking objectives. This formulation preserves gait topology while permitting independent end-effector adaptations for irregular terrains. We further introduce a terrain adaptation module with a dynamic replay buffer to resolve the policy's distribution shifts across different terrains. We validate our method across four locomotion styles and four terrains, demonstrating that LatentMimic enables effective terrain-adaptive locomotion, achieving higher terrain traversal success rates than state-of-the-art motion-tracking methods while maintaining high stylistic fidelity.

  PublisherOA PDF

• Game-Based and Gamified Robotics Education: A Comparative Systematic Review and Design Guidelines

  2026-04-13 · 1 citations

  articleOpen access

  Robotics education fosters computational thinking, creativity, and problem-solving, but remains challenging due to technical complexity. Game-based learning (GBL) and gamification offer engagement benefits, yet their comparative impact remains unclear. We present the first PRISMA-aligned systematic review and comparative synthesis of GBL and gamification in robotics education, analyzing 95 studies from 12,485 records across four databases (2014–2025). We coded each study’s approach, learning context, skill level, modality, pedagogy, and outcomes (κ =.918). Three patterns emerged: (1) approach–context–pedagogy coupling (GBL more prevalent in informal settings, while gamification dominated formal classrooms [p <.001] and favored project-based learning [p =.009]); (2) emphasis on introductory programming and modular kits, with limited adoption of advanced software (~17%), advanced hardware (~5%), or immersive technologies (~22%); and (3) short study horizons, relying on self-report. We propose eight research directions and a design space outlining best practices and pitfalls, offering actionable guidance for robotics education.

  PublisherDOI

• Tuning-Free Amodal Segmentation via the Occlusion-Free Bias of Inpainting Models

  Proceedings of the AAAI Conference on Artificial Intelligence · 2026-03-14

  articleOpen access

  Amodal segmentation is an image-based algorithm that aims to predict masks for both visible and occluded parts of objects. Existing methods typically rely on supervised learning with annotated amodal masks or synthetic data. The effectiveness of these methods relies heavily on the quality of the datasets. This dependence can unintentionally restrict their generalization capabilities due to insufficient diversity and size. Although existing zero-shot methods perform well on their reported datasets, their performance does not necessarily transfer to other datasets. We propose a tuning-free approach that re-purposes diffusion-based inpainting foundation models for amodal segmentation. Our approach is motivated by the “occlusion-free bias” of inpainting models, i.e., the inpainted objects tend to be complete and without occlusions. We reconstruct the occluded regions of an object via inpainting and then apply segmentation, all without additional training or fine-tuning. Experiments on five datasets, three previously unreported, demonstrate the generalizability of our approach. On average, our approach achieves 5.3% more accurate masks in mIoU compared to the publicly available state-of-the-art, pix2gestalt.

  PublisherDOI

• LatentMimic: Terrain-Adaptive Locomotion via Latent Space Imitation

  arXiv (Cornell University) · 2026-04-07

  preprintOpen accessSenior author

  Developing natural and diverse locomotion controllers for quadruped robots that can adapt to complex terrains while preserving motion style remains a significant challenge. Existing imitation-based methods face a fundamental optimization trade-off: strict adherence to motion capture (mocap) references penalizes the geometric deviations required for terrain adaptability, whereas terrain-centric policies often compromise stylistic fidelity. We introduce LatentMimic, a novel locomotion learning framework that decouples stylistic fidelity from geometric constraints. By minimizing the marginal latent divergence between the policy's state-action distribution and a learned mocap prior, our approach provides a conditional relaxation of rigid pose-tracking objectives. This formulation preserves gait topology while permitting independent end-effector adaptations for irregular terrains. We further introduce a terrain adaptation module with a dynamic replay buffer to resolve the policy's distribution shifts across different terrains. We validate our method across four locomotion styles and four terrains, demonstrating that LatentMimic enables effective terrain-adaptive locomotion, achieving higher terrain traversal success rates than state-of-the-art motion-tracking methods while maintaining high stylistic fidelity.

  PublisherDOI

Recent grants

• EAGER: Collaborative Research: Inverse Procedural Material Modeling for Battery Design

  NSF · $150k · 2017–2018

• Collaborative Research: TRTech-PGR: Digital Ideotype for Optimal Canopy Architecture

  NSF · $533k · 2024–2027

• Collaborative Research: IIBRP: VR-Bio-Talk: VR Voice-Controlled Visual Analytics Platform for Plant Digital Twins

  NSF · $1.2M · 2024–2027

Frequent coauthors

• Éric Galin

  Laboratoire d'Informatique en Images et Systèmes d'Information

  48 shared

• Éric Guérin

  Centre National de la Recherche Scientifique

  39 shared

• Adrien Peytavie

  Centre National de la Recherche Scientifique

  36 shared

• Daniel G. Aliaga

  Purdue University West Lafayette

  27 shared

• Alejandra J. Magana

  Bridge University

  25 shared

• Sören Pirk

  21 shared

• Marie‐Paule Cani

  Centre National de la Recherche Scientifique

  20 shared

• Vojtĕch Krs

  Adobe Systems (United States)

  19 shared

Education

• PhD, Computer Science

  České vysoké učení technické v Praze

  1998

• MS, Computer Science

  České vysoké učení technické v Praze

  1991

Awards & honors

• Eurographics Fellow
• Senior member of ACM and IEEE
• Editor-in-chief of Elsevier's Graphical Models