Search Results for author: Peter Wonka

Found 83 papers, 37 papers with code

Functional Diffusion

no code implementations26 Nov 2023 Biao Zhang, Peter Wonka

We propose a new class of generative diffusion models, called functional diffusion.

3DCoMPaT$^{++}$: An improved Large-scale 3D Vision Dataset for Compositional Recognition

no code implementations27 Oct 2023 Habib Slim, Xiang Li, Yuchen Li, Mahmoud Ahmed, Mohamed Ayman, Ujjwal Upadhyay, Ahmed Abdelreheem, Arpit Prajapati, Suhail Pothigara, Peter Wonka, Mohamed Elhoseiny

In this work, we present 3DCoMPaT$^{++}$, a multimodal 2D/3D dataset with 160 million rendered views of more than 10 million stylized 3D shapes carefully annotated at the part-instance level, alongside matching RGB point clouds, 3D textured meshes, depth maps, and segmentation masks.

LLM Blueprint: Enabling Text-to-Image Generation with Complex and Detailed Prompts

1 code implementation16 Oct 2023 Hanan Gani, Shariq Farooq Bhat, Muzammal Naseer, Salman Khan, Peter Wonka

Diffusion-based generative models have significantly advanced text-to-image generation but encounter challenges when processing lengthy and intricate text prompts describing complex scenes with multiple objects.

Layout-to-Image Generation Scene Generation

State of the Art on Diffusion Models for Visual Computing

no code implementations11 Oct 2023 Ryan Po, Wang Yifan, Vladislav Golyanik, Kfir Aberman, Jonathan T. Barron, Amit H. Bermano, Eric Ryan Chan, Tali Dekel, Aleksander Holynski, Angjoo Kanazawa, C. Karen Liu, Lingjie Liu, Ben Mildenhall, Matthias Nießner, Björn Ommer, Christian Theobalt, Peter Wonka, Gordon Wetzstein

The field of visual computing is rapidly advancing due to the emergence of generative artificial intelligence (AI), which unlocks unprecedented capabilities for the generation, editing, and reconstruction of images, videos, and 3D scenes.

WinSyn: A High Resolution Testbed for Synthetic Data

1 code implementation9 Oct 2023 Tom Kelly, John Femiani, Peter Wonka

We argue that the dataset is a crucial step to enable future research in synthetic data generation for deep learning.

Synthetic Data Generation

Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors

1 code implementation30 Jun 2023 Guocheng Qian, Jinjie Mai, Abdullah Hamdi, Jian Ren, Aliaksandr Siarohin, Bing Li, Hsin-Ying Lee, Ivan Skorokhodov, Peter Wonka, Sergey Tulyakov, Bernard Ghanem

We present Magic123, a two-stage coarse-to-fine approach for high-quality, textured 3D meshes generation from a single unposed image in the wild using both2D and 3D priors.

Image to 3D

Zero-Shot 3D Shape Correspondence

no code implementations5 Jun 2023 Ahmed Abdelreheem, Abdelrahman Eldesokey, Maks Ovsjanikov, Peter Wonka

Instead, we propose to exploit the in-context learning capabilities of ChatGPT to generate two different sets of semantic regions for each shape and a semantic mapping between them.

PET-NeuS: Positional Encoding Tri-Planes for Neural Surfaces

1 code implementation CVPR 2023 Yiqun Wang, Ivan Skorokhodov, Peter Wonka

The first component is to borrow the tri-plane representation from EG3D and represent signed distance fields as a mixture of tri-planes and MLPs instead of representing it with MLPs only.

Surface Reconstruction

SATR: Zero-Shot Semantic Segmentation of 3D Shapes

no code implementations ICCV 2023 Ahmed Abdelreheem, Ivan Skorokhodov, Maks Ovsjanikov, Peter Wonka

We explore the task of zero-shot semantic segmentation of 3D shapes by using large-scale off-the-shelf 2D image recognition models.

Segmentation Semantic Segmentation +1

VIVE3D: Viewpoint-Independent Video Editing using 3D-Aware GANs

1 code implementation CVPR 2023 Anna Frühstück, Nikolaos Sarafianos, Yuanlu Xu, Peter Wonka, Tony Tung

Our experiments demonstrate that VIVE3D generates high-fidelity face edits at consistent quality from a range of camera viewpoints which are composited with the original video in a temporally and spatially consistent manner.

Optical Flow Estimation Video Editing

3D generation on ImageNet

no code implementations2 Mar 2023 Ivan Skorokhodov, Aliaksandr Siarohin, Yinghao Xu, Jian Ren, Hsin-Ying Lee, Peter Wonka, Sergey Tulyakov

Existing 3D-from-2D generators are typically designed for well-curated single-category datasets, where all the objects have (approximately) the same scale, 3D location, and orientation, and the camera always points to the center of the scene.

ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth

3 code implementations23 Feb 2023 Shariq Farooq Bhat, Reiner Birkl, Diana Wofk, Peter Wonka, Matthias Müller

Finally, ZoeD-M12-NK is the first model that can jointly train on multiple datasets (NYU Depth v2 and KITTI) without a significant drop in performance and achieve unprecedented zero-shot generalization performance to eight unseen datasets from both indoor and outdoor domains.

Ranked #5 on Monocular Depth Estimation on NYU-Depth V2 (using extra training data)

Monocular Depth Estimation

3DAvatarGAN: Bridging Domains for Personalized Editable Avatars

no code implementations CVPR 2023 Rameen Abdal, Hsin-Ying Lee, Peihao Zhu, Menglei Chai, Aliaksandr Siarohin, Peter Wonka, Sergey Tulyakov

Finally, we propose a novel inversion method for 3D-GANs linking the latent spaces of the source and the target domains.

GPR-Net: Multi-view Layout Estimation via a Geometry-aware Panorama Registration Network

no code implementations20 Oct 2022 Jheng-Wei Su, Chi-Han Peng, Peter Wonka, Hung-Kuo Chu

The major improvement over PSMNet comes from a novel Geometry-aware Panorama Registration Network or GPR-Net that effectively tackles the wide baseline registration problem by exploiting the layout geometry and computing fine-grained correspondences on the layout boundaries, instead of the global pixel-space.


Large-Scale Auto-Regressive Modeling Of Street Networks

1 code implementation1 Sep 2022 Michael Birsak, Tom Kelly, Wamiq Para, Peter Wonka

We present a novel generative method for the creation of city-scale road layouts.

Learning to Construct 3D Building Wireframes from 3D Line Clouds

1 code implementation25 Aug 2022 Yicheng Luo, Jing Ren, Xuefei Zhe, Di Kang, Yajing Xu, Peter Wonka, Linchao Bao

The network takes a line cloud as input , i. e., a nonstructural and unordered set of 3D line segments extracted from multi-view images, and outputs a 3D wireframe of the underlying building, which consists of a sparse set of 3D junctions connected by line segments.

EpiGRAF: Rethinking training of 3D GANs

1 code implementation21 Jun 2022 Ivan Skorokhodov, Sergey Tulyakov, Yiqun Wang, Peter Wonka

In this work, we show that it is possible to obtain a high-resolution 3D generator with SotA image quality by following a completely different route of simply training the model patch-wise.

3D-Aware Image Synthesis

Gaussian Blue Noise

no code implementations15 Jun 2022 Abdalla G. M. Ahmed, Jing Ren, Peter Wonka

Among the various approaches for producing point distributions with blue noise spectrum, we argue for an optimization framework using Gaussian kernels.

HF-NeuS: Improved Surface Reconstruction Using High-Frequency Details

1 code implementation15 Jun 2022 Yiqun Wang, Ivan Skorokhodov, Peter Wonka

We develop HF-NeuS, a novel method to improve the quality of surface reconstruction in neural rendering.

Neural Rendering Surface Reconstruction +1

COFS: Controllable Furniture layout Synthesis

no code implementations29 May 2022 Wamiq Reyaz Para, Paul Guerrero, Niloy Mitra, Peter Wonka

Scalable generation of furniture layouts is essential for many applications in virtual reality, augmented reality, game development and synthetic data generation.

Language Modelling Synthetic Data Generation

3DILG: Irregular Latent Grids for 3D Generative Modeling

1 code implementation27 May 2022 Biao Zhang, Matthias Nießner, Peter Wonka

All probabilistic experiments confirm that we are able to generate detailed and high quality shapes to yield the new state of the art in generative 3D shape modeling.

3D Shape Modeling

Video2StyleGAN: Disentangling Local and Global Variations in a Video

no code implementations27 May 2022 Rameen Abdal, Peihao Zhu, Niloy J. Mitra, Peter Wonka

Image editing using a pretrained StyleGAN generator has emerged as a powerful paradigm for facial editing, providing disentangled controls over age, expression, illumination, etc.

Facial Editing

Differentiable Electron Microscopy Simulation: Methods and Applications for Visualization

no code implementations8 May 2022 Ngan Nguyen, Feng Liang, Dominik Engel, Ciril Bohak, Peter Wonka, Timo Ropinski, Ivan Viola

We propose a new microscopy simulation system that can depict atomistic models in a micrograph visual style, similar to results of physical electron microscopy imaging.


LocalBins: Improving Depth Estimation by Learning Local Distributions

1 code implementation28 Mar 2022 Shariq Farooq Bhat, Ibraheem Alhashim, Peter Wonka

We build on AdaBins which estimates a global distribution of depth values for the input image and evolve the architecture in two ways.

Monocular Depth Estimation

InsetGAN for Full-Body Image Generation

2 code implementations CVPR 2022 Anna Frühstück, Krishna Kumar Singh, Eli Shechtman, Niloy J. Mitra, Peter Wonka, Jingwan Lu

Instead of modeling this complex domain with a single GAN, we propose a novel method to combine multiple pretrained GANs, where one GAN generates a global canvas (e. g., human body) and a set of specialized GANs, or insets, focus on different parts (e. g., faces, shoes) that can be seamlessly inserted onto the global canvas.

Image Generation

On the Robustness of Quality Measures for GANs

1 code implementation31 Jan 2022 Motasem Alfarra, Juan C. Pérez, Anna Frühstück, Philip H. S. Torr, Peter Wonka, Bernard Ghanem

Finally, we show that the FID can be robustified by simply replacing the standard Inception with a robust Inception.

CLIP2StyleGAN: Unsupervised Extraction of StyleGAN Edit Directions

no code implementations9 Dec 2021 Rameen Abdal, Peihao Zhu, John Femiani, Niloy J. Mitra, Peter Wonka

The success of StyleGAN has enabled unprecedented semantic editing capabilities, on both synthesized and real images.

Zero-Shot Learning

Fast Sinkhorn Filters: Using Matrix Scaling for Non-Rigid Shape Correspondence With Functional Maps

no code implementations CVPR 2021 Gautam Pai, Jing Ren, Simone Melzi, Peter Wonka, Maks Ovsjanikov

In this paper, we provide a theoretical foundation for pointwise map recovery from functional maps and highlight its relation to a range of shape correspondence methods based on spectral alignment.

Self-Supervised Learning of Domain Invariant Features for Depth Estimation

no code implementations4 Jun 2021 Hiroyasu Akada, Shariq Farooq Bhat, Ibraheem Alhashim, Peter Wonka

Specifically, we extend self-supervised learning from traditional representation learning, which works on images from a single domain, to domain invariant representation learning, which works on images from two different domains by utilizing an image-to-image translation network.

Depth Estimation Domain Adaptation +5

Barbershop: GAN-based Image Compositing using Segmentation Masks

1 code implementation2 Jun 2021 Peihao Zhu, Rameen Abdal, John Femiani, Peter Wonka

Seamlessly blending features from multiple images is extremely challenging because of complex relationships in lighting, geometry, and partial occlusion which cause coupling between different parts of the image.

Finding Nano-Ötzi: Semi-Supervised Volume Visualization for Cryo-Electron Tomography

no code implementations4 Apr 2021 Ngan Nguyen, Ciril Bohak, Dominik Engel, Peter Mindek, Ondřej Strnad, Peter Wonka, Sai Li, Timo Ropinski, Ivan Viola

Our technique shows the high impact in target sciences for visual data analysis of very noisy volumes that cannot be visualized with existing techniques.

Electron Tomography Segmentation

IntraTomo: Self-Supervised Learning-Based Tomography via Sinogram Synthesis and Prediction

1 code implementation ICCV 2021 Guangming Zang, Ramzi Idoughi, Rui Li, Peter Wonka, Wolfgang Heidrich

After getting estimated through the sinogram prediction module, the density field is consistently refined in the second module using local and non-local geometrical priors.

Computed Tomography (CT) Low-Dose X-Ray Ct Reconstruction +3

Improved StyleGAN Embedding: Where are the Good Latents?

3 code implementations13 Dec 2020 Peihao Zhu, Rameen Abdal, Yipeng Qin, John Femiani, Peter Wonka

First, we introduce a new normalized space to analyze the diversity and the quality of the reconstructed latent codes.

Training Data Generating Networks: Shape Reconstruction via Bi-level Optimization

no code implementations ICLR 2022 Biao Zhang, Peter Wonka

We combine training data generating networks with bi-level optimization algorithms to obtain a complete framework for which all components can be jointly trained.

3D Shape Reconstruction 3D Shape Representation +2

Channel-Directed Gradients for Optimization of Convolutional Neural Networks

no code implementations25 Aug 2020 Dong Lao, Peihao Zhu, Peter Wonka, Ganesh Sundaramoorthi

We introduce optimization methods for convolutional neural networks that can be used to improve existing gradient-based optimization in terms of generalization error.

StyleFlow: Attribute-conditioned Exploration of StyleGAN-Generated Images using Conditional Continuous Normalizing Flows

3 code implementations6 Aug 2020 Rameen Abdal, Peihao Zhu, Niloy Mitra, Peter Wonka

We evaluate our method using the face and the car latent space of StyleGAN, and demonstrate fine-grained disentangled edits along various attributes on both real photographs and StyleGAN generated images.

Disentangled Image Generation Through Structured Noise Injection

1 code implementation CVPR 2020 Yazeed Alharbi, Peter Wonka

We show that disentanglement in the first layer of the generator network leads to disentanglement in the generated image.

Disentanglement Image Generation

MGCN: Descriptor Learning using Multiscale GCNs

no code implementations28 Jan 2020 Yiqun Wang, Jing Ren, Dong-Ming Yan, Jianwei Guo, Xiaopeng Zhang, Peter Wonka

Second, we propose a new multiscale graph convolutional network (MGCN) to transform a non-learned feature to a more discriminative descriptor.

SEAN: Image Synthesis with Semantic Region-Adaptive Normalization

1 code implementation CVPR 2020 Peihao Zhu, Rameen Abdal, Yipeng Qin, Peter Wonka

Using SEAN normalization, we can build a network architecture that can control the style of each semantic region individually, e. g., we can specify one style reference image per region.

Image Generation Segmentation

StructEdit: Learning Structural Shape Variations

1 code implementation CVPR 2020 Kaichun Mo, Paul Guerrero, Li Yi, Hao Su, Peter Wonka, Niloy Mitra, Leonidas J. Guibas

Learning to encode differences in the geometry and (topological) structure of the shapes of ordinary objects is key to generating semantically plausible variations of a given shape, transferring edits from one shape to another, and many other applications in 3D content creation.

Photorealistic Material Editing Through Direct Image Manipulation

no code implementations12 Sep 2019 Károly Zsolnai-Fehér, Peter Wonka, Michael Wimmer

In the proposed workflow, the user starts with an input image and applies a few intuitive transforms (e. g., colorization, image inpainting) within a 2D image editor of their choice, and in the next step, our technique produces a photorealistic result that approximates this target image.

Colorization Image Inpainting +1

StructureNet: Hierarchical Graph Networks for 3D Shape Generation

2 code implementations1 Aug 2019 Kaichun Mo, Paul Guerrero, Li Yi, Hao Su, Peter Wonka, Niloy Mitra, Leonidas J. Guibas

We introduce StructureNet, a hierarchical graph network which (i) can directly encode shapes represented as such n-ary graphs; (ii) can be robustly trained on large and complex shape families; and (iii) can be used to generate a great diversity of realistic structured shape geometries.

3D Shape Generation

TileGAN: Synthesis of Large-Scale Non-Homogeneous Textures

1 code implementation29 Apr 2019 Anna Frühstück, Ibraheem Alhashim, Peter Wonka

We tackle the problem of texture synthesis in the setting where many input images are given and a large-scale output is required.

Image Generation Image Stylization +1

ZoomOut: Spectral Upsampling for Efficient Shape Correspondence

2 code implementations16 Apr 2019 Simone Melzi, Jing Ren, Emanuele Rodolà, Abhishek Sharma, Peter Wonka, Maks Ovsjanikov

Our main observation is that high quality maps can be obtained even if the input correspondences are noisy or are encoded by a small number of coefficients in a spectral basis.


High Quality Monocular Depth Estimation via Transfer Learning

43 code implementations31 Dec 2018 Ibraheem Alhashim, Peter Wonka

Accurate depth estimation from images is a fundamental task in many applications including scene understanding and reconstruction.

Action Classification Monocular Depth Estimation +2

Latent Filter Scaling for Multimodal Unsupervised Image-to-Image Translation

no code implementations CVPR 2019 Yazeed Alharbi, Neil Smith, Peter Wonka

In multimodal unsupervised image-to-image translation tasks, the goal is to translate an image from the source domain to many images in the target domain.

Disentanglement Multimodal Unsupervised Image-To-Image Translation +2

How does Lipschitz Regularization Influence GAN Training?

no code implementations ECCV 2020 Yipeng Qin, Niloy Mitra, Peter Wonka

In this work, we uncover an even more important effect of Lipschitz regularization by examining its impact on the loss function: It degenerates GAN loss functions to almost linear ones by restricting their domain and interval of attainable gradient values.

Super-Resolution and Sparse View CT Reconstruction

no code implementations ECCV 2018 Guangming Zang, Mohamed Aly, Ramzi Idoughi, Peter Wonka, Wolfgang Heidrich

As a second, smaller contribution, we also show that when using such a proximal reconstruction framework, it is beneficial to employ the Simultaneous Algebraic Reconstruction Technique (SART) instead of the commonly used Conjugate Gradient (CG) method in the solution of the data term proximal operator.

Computed Tomography (CT) Super-Resolution

FrankenGAN: Guided Detail Synthesis for Building Mass-Models Using Style-Synchonized GANs

2 code implementations19 Jun 2018 Tom Kelly, Paul Guerrero, Anthony Steed, Peter Wonka, Niloy J. Mitra

The various GANs are synchronized to produce consistent style distributions over buildings and neighborhoods.


Facade Segmentation in the Wild

no code implementations9 May 2018 John Femiani, Wamiq Reyaz Para, Niloy Mitra, Peter Wonka

Specifically, we propose a MULTIFACSEGNET architecture to assign multiple labels to each pixel, a SEPARABLE architecture as a low-rank formulation that encourages extraction of rectangular elements, and a COMPATIBILITY network that simultaneously seeks segmentation across facade element types allowing the network to 'see' intermediate output probabilities of the various facade element classes.

Image Segmentation Segmentation +1

Gaussian Material Synthesis

no code implementations23 Apr 2018 Károly Zsolnai-Fehér, Peter Wonka, Michael Wimmer

Workflow timings against Disney's "principled" shader reveal that our system scales well with the number of sought materials, thus empowering even novice users to generate hundreds of high-quality material models without any expertise in material modeling.

Supervised Convolutional Sparse Coding

no code implementations8 Apr 2018 Lama Affara, Bernard Ghanem, Peter Wonka

Convolutional Sparse Coding (CSC) is a well-established image representation model especially suited for image restoration tasks.

Image Reconstruction Image Restoration

Integration of Absolute Orientation Measurements in the KinectFusion Reconstruction pipeline

no code implementations12 Feb 2018 Silvio Giancola, Jens Schneider, Peter Wonka, Bernard S. Ghanem

We also present a technique to filter the pairs of 3D matched points based on the distribution of their distances.

3D Reconstruction

Fast Convolutional Sparse Coding in the Dual Domain

no code implementations27 Sep 2017 Lama Affara, Bernard Ghanem, Peter Wonka

Convolutional sparse coding (CSC) is an important building block of many computer vision applications ranging from image and video compression to deep learning.

Video Compression

TRex: A Tomography Reconstruction Proximal Framework for Robust Sparse View X-Ray Applications

1 code implementation11 Jun 2016 Mohamed Aly, Guangming Zang, Wolfgang Heidrich, Peter Wonka

We present TRex, a flexible and robust Tomographic Reconstruction framework using proximal algorithms.

Intrinsic Scene Decomposition From RGB-D images

no code implementations ICCV 2015 Mohammed Hachama, Bernard Ghanem, Peter Wonka

In this paper, we address the problem of computing an intrinsic decomposition of the colors of a surface into an albedo and a shading term.

Intrinsic Image Decomposition

RAID: A Relation-Augmented Image Descriptor

no code implementations5 Oct 2015 Paul Guerrero, Niloy J. Mitra, Peter Wonka

As humans, we regularly interpret images based on the relations between image regions.

Scaling SVM and Least Absolute Deviations via Exact Data Reduction

no code implementations25 Oct 2013 Jie Wang, Peter Wonka, Jieping Ye

Some appealing features of our screening method are: (1) DVI is safe in the sense that the vectors discarded by DVI are guaranteed to be non-support vectors; (2) the data set needs to be scanned only once to run the screening, whose computational cost is negligible compared to that of solving the SVM problem; (3) DVI is independent of the solvers and can be integrated with any existing efficient solvers.

A Safe Screening Rule for Sparse Logistic Regression

no code implementations NeurIPS 2014 Jie Wang, Jiayu Zhou, Jun Liu, Peter Wonka, Jieping Ye

The l1-regularized logistic regression (or sparse logistic regression) is a widely used method for simultaneous classification and feature selection.

feature selection regression

Lasso Screening Rules via Dual Polytope Projection

no code implementations NeurIPS 2013 Jie Wang, Peter Wonka, Jieping Ye

To improve the efficiency of solving large-scale Lasso problems, El Ghaoui and his colleagues have proposed the SAFE rules which are able to quickly identify the inactive predictors, i. e., predictors that have $0$ components in the solution vector.

Fused Multiple Graphical Lasso

no code implementations10 Sep 2012 Sen Yang, Zhaosong Lu, Xiaotong Shen, Peter Wonka, Jieping Ye

We expect the two brain networks for NC and MCI to share common structures but not to be identical to each other; similarly for the two brain networks for MCI and AD.

Multi-Stage Dantzig Selector

no code implementations NeurIPS 2010 Ji Liu, Peter Wonka, Jieping Ye

We show that if $X$ obeys a certain condition, then with a large probability the difference between the solution $\hat\beta$ estimated by the proposed method and the true solution $\beta^*$ measured in terms of the $l_p$ norm ($p\geq 1$) is bounded as \begin{equation*} \|\hat\beta-\beta^*\|_p\leq \left(C(s-N)^{1/p}\sqrt{\log m}+\Delta\right)\sigma, \end{equation*} $C$ is a constant, $s$ is the number of nonzero entries in $\beta^*$, $\Delta$ is independent of $m$ and is much smaller than the first term, and $N$ is the number of entries of $\beta^*$ larger than a certain value in the order of $\mathcal{O}(\sigma\sqrt{\log m})$.

feature selection

Cannot find the paper you are looking for? You can Submit a new open access paper.