Bedrich Benes

Jae Joong Lee, Bosheng Li, Sara Beery et al. · mit

We introduce Tree D-fusion, featuring the first collection of 600,000 environmentally aware, 3D simulation-ready tree models generated through Diffusion priors. Each reconstructed 3D tree model corresponds to an image from Google's Auto Arborist Dataset, comprising street view images and associated genus labels of trees across North America. Our method distills the scores of two tree-adapted diffusion models by utilizing text prompts to specify a tree genus, thus facilitating shape reconstruction. This process involves reconstructing a 3D tree envelope filled with point markers, which are subsequently utilized to estimate the tree's branching structure using the space colonization algorithm conditioned on a specified genus.

Yichen Sheng, Yifan Liu, Jianming Zhang et al.

Shadows are essential for realistic image compositing. Physics-based shadow rendering methods require 3D geometries, which are not always available. Deep learning-based shadow synthesis methods learn a mapping from the light information to an object's shadow without explicitly modeling the shadow geometry. Still, they lack control and are prone to visual artifacts. We introduce pixel heigh, a novel geometry representation that encodes the correlations between objects, ground, and camera pose. The pixel height can be calculated from 3D geometries, manually annotated on 2D images, and can also be predicted from a single-view RGB image by a supervised approach. It can be used to calculate hard shadows in a 2D image based on the projective geometry, providing precise control of the shadows' direction and shape. Furthermore, we propose a data-driven soft shadow generator to apply softness to a hard shadow based on a softness input parameter. Qualitative and quantitative evaluations demonstrate that the proposed pixel height significantly improves the quality of the shadow generation while allowing for controllability.

Yichen Sheng, Jianming Zhang, Julien Philip et al. · pku

Lighting effects such as shadows or reflections are key in making synthetic images realistic and visually appealing. To generate such effects, traditional computer graphics uses a physically-based renderer along with 3D geometry. To compensate for the lack of geometry in 2D Image compositing, recent deep learning-based approaches introduced a pixel height representation to generate soft shadows and reflections. However, the lack of geometry limits the quality of the generated soft shadows and constrain reflections to pure specular ones. We introduce PixHt-Lab, a system leveraging an explicit mapping from pixel height representation to 3D space. Using this mapping, PixHt-Lab reconstructs both the cutout and background geometry and renders realistic, diverse, lighting effects for image compositing. Given a surface with physically-based materials, we can render reflections with varying glossiness. To generate more realistic soft shadows, we further propose to use 3D-aware buffer channels to guide a neural renderer. Both quantitative and qualitative evaluations demonstrate that PixHt-Lab significantly improves soft shadow generation.

Zhiquan Wang, Bedrich Benes, Ahmed H. Qureshi et al.

We introduce a novel co-design method for autonomous moving agents' shape attributes and locomotion by combining deep reinforcement learning and evolution with user control. Our main inspiration comes from evolution, which has led to wide variability and adaptation in Nature and has the potential to significantly improve design and behavior simultaneously. Our method takes an input agent with optional simple constraints such as leg parts that should not evolve or allowed ranges of changes. It uses physics-based simulation to determine its locomotion and finds a behavior policy for the input design, later used as a baseline for comparison. The agent is then randomly modified within the allowed ranges creating a new generation of several hundred agents. The generation is trained by transferring the previous policy, which significantly speeds up the training. The best-performing agents are selected, and a new generation is formed using their crossover and mutations. The next generations are then trained until satisfactory results are reached. We show a wide variety of evolved agents, and our results show that even with only 10% of changes, the overall performance of the evolved agents improves 50%. If more significant changes to the initial design are allowed, our experiments' performance improves even more to 150%. Contrary to related work, our co-design works on a single GPU and provides satisfactory results by training thousands of agents within one hour.

Simon Perche, Adrien Peytavie, Bedrich Benes et al.

Many terrain modelling methods have been proposed for the past decades, providing efficient and often interactive authoring tools. However, they generally do not include any notion of style, which is a critical aspect for designers in the entertainment industry. We introduce StyleDEM, a new generative adversarial network method for terrain synthesis and authoring, with a versatile toolbox of authoring methods with style. This method starts from an input sketch or an existing terrain. It outputs a terrain with features that can be authored using interactive brushes and enhanced with additional tools such as style manipulation or super-resolution. The strength of our approach resides in the versatility and interoperability of the toolbox.

Jae Joong Lee, Bedrich Benes

We introduce RGB2Point, an unposed single-view RGB image to a 3D point cloud generation based on Transformer. RGB2Point takes an input image of an object and generates a dense 3D point cloud. Contrary to prior works based on CNN layers and diffusion denoising approaches, we use pre-trained Transformer layers that are fast and generate high-quality point clouds with consistent quality over available categories. Our generated point clouds demonstrate high quality on a real-world dataset, as evidenced by improved Chamfer distance (51.15%) and Earth Mover's distance (45.96%) metrics compared to the current state-of-the-art. Additionally, our approach shows a better quality on a synthetic dataset, achieving better Chamfer distance (39.26%), Earth Mover's distance (26.95%), and F-score (47.16%). Moreover, our method produces 63.1% more consistent high-quality results across various object categories compared to prior works. Furthermore, RGB2Point is computationally efficient, requiring only 2.3GB of VRAM to reconstruct a 3D point cloud from a single RGB image, and our implementation generates the results 15,133x faster than a SOTA diffusion-based model.

Zhiquan Wang, Yunyu Liu, Dipam Patel et al.

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.

Amber Yijia Zheng, Jae Joong Lee, Bedrich Benes et al.

We present a vision-language model (VLM) that automatically edits website HTML to address Web Content Accessibility Guidelines 2 (WCAG2) violations. We formulate this as a supervised image-conditioned program synthesis task, where the model learns to correct HTML given the HTML and its rendering. We collected WebAccessVL, a new dataset with manually corrected accessibility violations, establishing paired training data. We then propose a violation-conditioned VLM that additionally conditions on the WCAG2 violation count to guide the correction process. Experiments demonstrate that our method effectively reduces the average number of violations from 5.34 to 0.44 per website, outperforming commercial LLM APIs (Gemini, GPT-5). A perceptual study confirms that our edited websites maintain the original visual appearance and content.

Jae Joong Lee, Bedrich Benes

Generating ground-level images from aerial views is a challenging task due to extreme viewpoint disparity, occlusions, and a limited field of view. We introduce Top2Ground, a novel diffusion-based method that directly generates photorealistic ground-view images from aerial input images without relying on intermediate representations such as depth maps or 3D voxels. Specifically, we condition the denoising process on a joint representation of VAE-encoded spatial features (derived from aerial RGB images and an estimated height map) and CLIP-based semantic embeddings. This design ensures the generation is both geometrically constrained by the scene's 3D structure and semantically consistent with its content. We evaluate Top2Ground on three diverse datasets: CVUSA, CVACT, and the Auto Arborist. Our approach shows 7.3% average improvement in SSIM across three benchmark datasets, showing Top2Ground can robustly handle both wide and narrow fields of view, highlighting its strong generalization capabilities.

Syed T. Mubarrat, Byung-Cheol Min, Tianyu Shao et al.

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 (k = .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.

Jae Joong Lee, Bedrich Benes

Deep learning-based 3D object reconstruction has achieved unprecedented results. Among those, the transformer deep neural model showed outstanding performance in many applications of computer vision. We introduce SnakeVoxFormer, a novel, 3D object reconstruction in voxel space from a single image using the transformer. The input to SnakeVoxFormer is a 2D image, and the result is a 3D voxel model. The key novelty of our approach is in using the run-length encoding that traverses (like a snake) the voxel space and encodes wide spatial differences into a 1D structure that is suitable for transformer encoding. We then use dictionary encoding to convert the discovered RLE blocks into tokens that are used for the transformer. The 1D representation is a lossless 3D shape data compression method that converts to 1D data that use only about 1% of the original data size. We show how different voxel traversing strategies affect the effect of encoding and reconstruction. We compare our method with the state-of-the-art for 3D voxel reconstruction from images and our method improves the state-of-the-art methods by at least 2.8% and up to 19.8%.

Zhiquan Wang, Bedrich Benes

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.

Lu Ling, Yichen Sheng, Zhi Tu et al.

We have witnessed significant progress in deep learning-based 3D vision, ranging from neural radiance field (NeRF) based 3D representation learning to applications in novel view synthesis (NVS). However, existing scene-level datasets for deep learning-based 3D vision, limited to either synthetic environments or a narrow selection of real-world scenes, are quite insufficient. This insufficiency not only hinders a comprehensive benchmark of existing methods but also caps what could be explored in deep learning-based 3D analysis. To address this critical gap, we present DL3DV-10K, a large-scale scene dataset, featuring 51.2 million frames from 10,510 videos captured from 65 types of point-of-interest (POI) locations, covering both bounded and unbounded scenes, with different levels of reflection, transparency, and lighting. We conducted a comprehensive benchmark of recent NVS methods on DL3DV-10K, which revealed valuable insights for future research in NVS. In addition, we have obtained encouraging results in a pilot study to learn generalizable NeRF from DL3DV-10K, which manifests the necessity of a large-scale scene-level dataset to forge a path toward a foundation model for learning 3D representation. Our DL3DV-10K dataset, benchmark results, and models will be publicly accessible at https://dl3dv-10k.github.io/DL3DV-10K/.

Jacek Kałużny, Yannik Schreckenberg, Karol Cyganik et al.

We introduce LAESI, a Synthetic Leaf Dataset of 100,000 synthetic leaf images on millimeter paper, each with semantic masks and surface area labels. This dataset provides a resource for leaf morphology analysis primarily aimed at beech and oak leaves. We evaluate the applicability of the dataset by training machine learning models for leaf surface area prediction and semantic segmentation, using real images for validation. Our validation shows that these models can be trained to predict leaf surface area with a relative error not greater than an average human annotator. LAESI also provides an efficient framework based on 3D procedural models and generative AI for the large-scale, controllable generation of data with potential further applications in agriculture and biology. We evaluate the inclusion of generative AI in our procedural data generation pipeline and show how data filtering based on annotation consistency results in datasets which allow training the highest performing vision models.

Jorge Askur Vazquez Fernandez, Jae Joong Lee, Santiago Andrés Serrano Vacca et al.

The paper introduces Hands-Free VR, a voice-based natural-language interface for VR. The user gives a command using their voice, the speech audio data is converted to text using a speech-to-text deep learning model that is fine-tuned for robustness to word phonetic similarity and to spoken English accents, and the text is mapped to an executable VR command using a large language model that is robust to natural language diversity. Hands-Free VR was evaluated in a controlled within-subjects study (N = 22) that asked participants to find specific objects and to place them in various configurations. In the control condition participants used a conventional VR user interface to grab, carry, and position the objects using the handheld controllers. In the experimental condition participants used Hands-Free VR. The results confirm that: (1) Hands-Free VR is robust to spoken English accents, as for 20 of our participants English was not their first language, and to word phonetic similarity, correctly transcribing the voice command 96.71% of the time; (2) Hands-Free VR is robust to natural language diversity, correctly mapping the transcribed command to an executable command in 97.83% of the time; (3) Hands-Free VR had a significant efficiency advantage over the conventional VR interface in terms of task completion time, total viewpoint translation, total view direction rotation, and total left and right hand translations; (4) Hands-Free VR received high user preference ratings in terms of ease of use, intuitiveness, ergonomics, reliability, and desirability.

Mozhgan Hadadi, Talukder Z. Jubery, Patrick S. Schnable et al.

Accurate 3D plant models are crucial for computational phenotyping and physics-based simulation; however, current approaches face significant limitations. Learning-based reconstruction methods require extensive species-specific training data and lack editability. Procedural modeling offers parametric control but demands specialized expertise in geometric modeling and an in-depth understanding of complex procedural rules, making it inaccessible to domain scientists. We present FloraForge, an LLM-assisted framework that enables domain experts to generate biologically accurate, fully parametric 3D plant models through iterative natural language Plant Refinements (PR), minimizing programming expertise. Our framework leverages LLM-enabled co-design to refine Python scripts that generate parameterized plant geometries as hierarchical B-spline surface representations with botanical constraints with explicit control points and parametric deformation functions. This representation can be easily tessellated into polygonal meshes with arbitrary precision, ensuring compatibility with functional structural plant analysis workflows such as light simulation, computational fluid dynamics, and finite element analysis. We demonstrate the framework on maize, soybean, and mung bean, fitting procedural models to empirical point cloud data through manual refinement of the Plant Descriptor (PD), human-readable files. The pipeline generates dual outputs: triangular meshes for visualization and triangular meshes with additional parametric metadata for quantitative analysis. This approach uniquely combines LLM-assisted template creation, mathematically continuous representations enabling both phenotyping and rendering, and direct parametric control through PD. The framework democratizes sophisticated geometric modeling for plant science while maintaining mathematical rigor.

Jae Joong Lee, Bedrich Benes, Raymond A. Yeh

Amodal segmentation aims to predict segmentation masks for both the visible and occluded regions of an object. Most existing works formulate this as a supervised learning problem, requiring manually annotated amodal masks or synthetic training data. Consequently, their performance depends on the quality of the datasets, which often lack diversity and scale. This work introduces a tuning-free approach that repurposes pretrained diffusion-based inpainting 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 objects without occlusions. Specifically, 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 demonstrate the generalizability and robustness of our approach. On average, our approach achieves 5.3% more accurate masks over the state-of-the-art.

Xiaochen Zhou, Bosheng Li, Bedrich Benes et al.

In this paper, we propose DeepTree, a novel method for modeling trees based on learning developmental rules for branching structures instead of manually defining them. We call our deep neural model situated latent because its behavior is determined by the intrinsic state -- encoded as a latent space of a deep neural model -- and by the extrinsic (environmental) data that is situated as the location in the 3D space and on the tree structure. We use a neural network pipeline to train a situated latent space that allows us to locally predict branch growth only based on a single node in the branch graph of a tree model. We use this representation to progressively develop new branch nodes, thereby mimicking the growth process of trees. Starting from a root node, a tree is generated by iteratively querying the neural network on the newly added nodes resulting in the branching structure of the whole tree. Our method enables generating a wide variety of tree shapes without the need to define intricate parameters that control their growth and behavior. Furthermore, we show that the situated latents can also be used to encode the environmental response of tree models, e.g., when trees grow next to obstacles. We validate the effectiveness of our method by measuring the similarity of our tree models and by procedurally generated ones based on a number of established metrics for tree form.

Till Niese, Sören Pirk, Matthias Albrecht et al.

The placement of vegetation plays a central role in the realism of virtual scenes. We introduce procedural placement models (PPMs) for vegetation in urban layouts. PPMs are environmentally sensitive to city geometry and allow identifying plausible plant positions based on structural and functional zones in an urban layout. PPMs can either be directly used by defining their parameters or can be learned from satellite images and land register data. Together with approaches for generating buildings and trees, this allows us to populate urban landscapes with complex 3D vegetation. The effectiveness of our framework is shown through examples of large-scale city scenes and close-ups of individually grown tree models; we also validate it by a perceptual user study.

Yichen Sheng, Jianming Zhang, Bedrich Benes

We introduce an interactive Soft Shadow Network (SSN) to generates controllable soft shadows for image compositing. SSN takes a 2D object mask as input and thus is agnostic to image types such as painting and vector art. An environment light map is used to control the shadow's characteristics, such as angle and softness. SSN employs an Ambient Occlusion Prediction module to predict an intermediate ambient occlusion map, which can be further refined by the user to provides geometric cues to modulate the shadow generation. To train our model, we design an efficient pipeline to produce diverse soft shadow training data using 3D object models. In addition, we propose an inverse shadow map representation to improve model training. We demonstrate that our model produces realistic soft shadows in real-time. Our user studies show that the generated shadows are often indistinguishable from shadows calculated by a physics-based renderer and users can easily use SSN through an interactive application to generate specific shadow effects in minutes.

Hao Kang, Jianming Zhang, Haoxiang Li et al.

We introduce a novel framework for automatic capturing of human portraits. The framework allows the robot to follow a person to the desired location using a Person Re-identification model. When composing is activated, the robot attempts to adjust its position to form the view that can best match the given template image, and finally takes a photograph. A template image can be predicted dynamically using an off-the-shelf photo evaluation model by the framework, or selected manually from a pre-defined set by the user. The template matching-based view adjustment is driven by a deep reinforcement learning network. Our framework lies on top of the Robot Operating System (ROS). The framework is designed to be modular so that all the models can be flexibly replaced based on needs. We show our framework on a variety of examples. In particular, we tested it in three indoor scenes and used it to take 20 photos of each scene: ten for the pre-defined template, ten for the dynamically generated ones. The average number of adjustment was $11.20$ for pre-defined templates and $12.76$ for dynamically generated ones; the average time spent was $22.11$ and $24.10$ seconds respectively.

Fanyou Wu, Rado Gazo, Eva Haviarova et al.

Recent advances show that deep neural networks are not robust to deliberately crafted adversarial examples which many are generated by adding human imperceptible perturbation to clear input. Consider $l_2$ norms attacks, Project Gradient Descent (PGD) and the Carlini and Wagner (C\&W) attacks are the two main methods, where PGD control max perturbation for adversarial examples while C\&W approach treats perturbation as a regularization term optimized it with loss function together. If we carefully set parameters for any individual input, both methods become similar. In general, PGD attacks perform faster but obtains larger perturbation to find adversarial examples than the C\&W when fixing the parameters for all inputs. In this report, we propose an efficient modified PGD method for attacking ensemble models by automatically changing ensemble weights and step size per iteration per input. This method generates smaller perturbation adversarial examples than PGD method while remains efficient as compared to C\&W method. Our method won the first place in IJCAI19 Targeted Adversarial Attack competition.

Sören Pirk, Vojtech Krs, Kaimo Hu et al.

Interactions play a key role in understanding objects and scenes, for both virtual and real world agents. We introduce a new general representation for proximal interactions among physical objects that is agnostic to the type of objects or interaction involved. The representation is based on tracking particles on one of the participating objects and then observing them with sensors appropriately placed in the interaction volume or on the interaction surfaces. We show how to factorize these interaction descriptors and project them into a particular participating object so as to obtain a new functional descriptor for that object, its interaction landscape, capturing its observed use in a spatio-temporal framework. Interaction landscapes are independent of the particular interaction and capture subtle dynamic effects in how objects move and behave when in functional use. Our method relates objects based on their function, establishes correspondences between shapes based on functional key points and regions, and retrieves peer and partner objects with respect to an interaction.