BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Christian Rupprecht

Christian Rupprecht appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
Computer VisionMachine LearningArtificial Intelligenceeess.IVGraphicsNeural and Evolutionary Computingphysics.app-phphysics.opticsRobotics

Catalog footprint

What is connected

20works
9topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

20 published item(s)

preprint2026arXiv

Articraft: An Agentic System for Scalable Articulated 3D Asset Generation

A bottleneck in learning to understand articulated 3D objects is the lack of large and diverse datasets. In this paper, we propose to leverage large language models (LLMs) to close this gap and generate articulated assets at scale. We reduce the problem of generating an articulated 3D asset to that of writing a program that builds it. We then introduce a new agentic system, Articraft, that writes such programs automatically. We design a programmatic interface and harness to help the LLM do so effectively. The LLM writes code against a domain-specific SDK for defining parts, composing geometry, specifying joints, and writing tests to validate the resulting assets. The harness exposes a restricted workspace and interface to the LLM, validates the resulting assets, and returns structured feedback. In this way, the LLM is not distracted by details such as authoring a URDF file or managing a complex software environment. We show that this produces higher-quality assets than both state-of-the-art articulated-asset generators and general-purpose coding agents. Using Articraft, we build Articraft-10K, a curated dataset of over 10K articulated assets spanning 245 categories, and show its utility both for training models of articulated assets and in downstream applications such as robotics simulation and virtual reality.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Probing into Camera Control of Video Models

Video is a rich and scalable source of 3D/4D visual observations, and camera control is a key capability for video generation models to produce geometrically meaningful content. Existing approaches typically learn a mapping from camera motion to video using additional camera modules and paired data. However, such datasets are often limited in scale, diversity, and scene dynamics, which can bias the model toward a narrow output distribution and compromise the strong prior learned by the base model. These limitations motivate a different perspective on camera control. In this paper, we show that camera control need not be modeled as an implicit mapping problem, but can instead be treated as a form of geometric guidance that induces displacements across frames. Specifically, we reformulate camera control into a set of displacement fields and apply them via differentiable resampling of latent features during denoising. Our simple approach achieves effective camera control with minimal degradation across diverse quality metrics compared to fine-tuned baselines. Since our method is applicable to most video diffusion models without training, it can also serve as a probe to study the camera control capabilities of base models. Using this probe, we identify universal biases shared by representative video models, as well as disparities in their responses to camera control. Finally, we benchmark their performance in multi-view generation, offering insights into their potential for 3D/4D tasks.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Syn4D: A Multiview Synthetic 4D Dataset

Dense 3D reconstruction and tracking of dynamic scenes from monocular video remains an important open challenge in computer vision. Progress in this area has been constrained by the scarcity of high-quality datasets with dense, complete, and accurate geometric annotations. To address this limitation, we introduce Syn4D, a multiview synthetic dataset of dynamic scenes that includes ground-truth camera motion, depth maps, dense tracking, and parametric human pose annotations. A key feature of Syn4D is the ability to unproject any pixel into 3D to any time and to any camera. We conduct extensive evaluations across multiple downstream tasks to demonstrate the utility and effectiveness of the proposed dataset, including 4D scene reconstruction, 3D point tracking, geometry-aware camera retargeting, and human pose estimation. The experimental results highlight Syn4D's potential to facilitate research in dynamic scene understanding and spatiotemporal modeling.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

VGGT-$Ω$

Recent feed-forward reconstruction models, such as VGGT, have proven competitive with traditional optimization-based reconstructors while also providing geometry-aware features useful for other tasks. Here, we show that the quality of these models scales predictably with model and data size. We do so by introducing VGGT-$Ω$, which substantially improves reconstruction accuracy, efficiency, and capabilities for both static and dynamic scenes. To enable training this model at an unprecedented scale, we introduce architectural changes that improve training efficiency, a high-quality data annotation pipeline that supports dynamic scenes, and a self-supervised learning protocol. We simplify VGGT's architecture by using a single dense prediction head with multi-task supervision and removing the expensive high-resolution convolutional layers. We also use registers to aggregate scene information into a compact representation and introduce register attention, which restricts inter-frame information exchange to these registers, in part replacing global attention. In this way, during training, VGGT-$Ω$ uses only about 30% of the GPU memory of its predecessor, allowing us to train with 15x more supervised data than prior work and to leverage vast amounts of unlabeled video data. VGGT-$Ω$ achieves strong results for reconstruction of static and dynamic scenes across multiple benchmarks, for example, improving over the previous best camera estimation accuracy on Sintel by 77%. We also show that the learned registers can improve vision-language-action models and support alignment with language, suggesting that reconstruction can be a powerful and scalable proxy task for spatial understanding. Project Page: http://vggt-omega.github.io/

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localization

Unsupervised localization and segmentation are long-standing computer vision challenges that involve decomposing an image into semantically-meaningful segments without any labeled data. These tasks are particularly interesting in an unsupervised setting due to the difficulty and cost of obtaining dense image annotations, but existing unsupervised approaches struggle with complex scenes containing multiple objects. Differently from existing methods, which are purely based on deep learning, we take inspiration from traditional spectral segmentation methods by reframing image decomposition as a graph partitioning problem. Specifically, we examine the eigenvectors of the Laplacian of a feature affinity matrix from self-supervised networks. We find that these eigenvectors already decompose an image into meaningful segments, and can be readily used to localize objects in a scene. Furthermore, by clustering the features associated with these segments across a dataset, we can obtain well-delineated, nameable regions, i.e. semantic segmentations. Experiments on complex datasets (Pascal VOC, MS-COCO) demonstrate that our simple spectral method outperforms the state-of-the-art in unsupervised localization and segmentation by a significant margin. Furthermore, our method can be readily used for a variety of complex image editing tasks, such as background removal and compositing.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

DOVE: Learning Deformable 3D Objects by Watching Videos

Learning deformable 3D objects from 2D images is often an ill-posed problem. Existing methods rely on explicit supervision to establish multi-view correspondences, such as template shape models and keypoint annotations, which restricts their applicability on objects "in the wild". A more natural way of establishing correspondences is by watching videos of objects moving around. In this paper, we present DOVE, a method that learns textured 3D models of deformable object categories from monocular videos available online, without keypoint, viewpoint or template shape supervision. By resolving symmetry-induced pose ambiguities and leveraging temporal correspondences in videos, the model automatically learns to factor out 3D shape, articulated pose and texture from each individual RGB frame, and is ready for single-image inference at test time. In the experiments, we show that existing methods fail to learn sensible 3D shapes without additional keypoint or template supervision, whereas our method produces temporally consistent 3D models, which can be animated and rendered from arbitrary viewpoints.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Thermographic detection of internal defects using 2D photothermal super resolution reconstruction with sequential laser heating

Thermographic photothermal super resolution reconstruction enables the resolution of internal defects/inhomogeneities below the classical limit which is governed by the diffusion properties of thermal wave propagation. Based on a combination of the application of special sampling strategies and a subsequent numerical optimization step in post-processing, thermographic super resolution has already proven to be superior to standard thermographic methods in the detection of one-dimensional defect/inhomogeneity structures. In our work, we report an extension of the capabilities of the method for efficient detection and resolution of defect cross sections with fully two-dimensional structured laser-based heating. The reconstruction is carried out using one of two different algorithms which are proposed within this work. Both algorithms utilize the combination of several coherent measurements using convex optimization and exploit the sparse nature of defects/inhomogeneities as is typical for most nondestructive testing scenarios. Finally, the performance of each algorithm is rated on reconstruction quality and algorithmic complexity. The presented experimental approach is based on repeated spatially structured heating by a high power laser. As a result, a two-dimensional sparse defect/inhomogeneity map can be obtained. In addition, the obtained results are compared with those of conventional thermographic inspection methods which make use of homogeneous illumination. Due to the sparse nature of the reconstructed defect/inhomogeneity map, this comparison is performed qualitatively.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Unsupervised Part Discovery from Contrastive Reconstruction

The goal of self-supervised visual representation learning is to learn strong, transferable image representations, with the majority of research focusing on object or scene level. On the other hand, representation learning at part level has received significantly less attention. In this paper, we propose an unsupervised approach to object part discovery and segmentation and make three contributions. First, we construct a proxy task through a set of objectives that encourages the model to learn a meaningful decomposition of the image into its parts. Secondly, prior work argues for reconstructing or clustering pre-computed features as a proxy to parts; we show empirically that this alone is unlikely to find meaningful parts; mainly because of their low resolution and the tendency of classification networks to spatially smear out information. We suggest that image reconstruction at the level of pixels can alleviate this problem, acting as a complementary cue. Lastly, we show that the standard evaluation based on keypoint regression does not correlate well with segmentation quality and thus introduce different metrics, NMI and ARI, that better characterize the decomposition of objects into parts. Our method yields semantic parts which are consistent across fine-grained but visually distinct categories, outperforming the state of the art on three benchmark datasets. Code is available at the project page: https://www.robots.ox.ac.uk/~vgg/research/unsup-parts/.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Labelling unlabelled videos from scratch with multi-modal self-supervision

A large part of the current success of deep learning lies in the effectiveness of data -- more precisely: labelled data. Yet, labelling a dataset with human annotation continues to carry high costs, especially for videos. While in the image domain, recent methods have allowed to generate meaningful (pseudo-) labels for unlabelled datasets without supervision, this development is missing for the video domain where learning feature representations is the current focus. In this work, we a) show that unsupervised labelling of a video dataset does not come for free from strong feature encoders and b) propose a novel clustering method that allows pseudo-labelling of a video dataset without any human annotations, by leveraging the natural correspondence between the audio and visual modalities. An extensive analysis shows that the resulting clusters have high semantic overlap to ground truth human labels. We further introduce the first benchmarking results on unsupervised labelling of common video datasets Kinetics, Kinetics-Sound, VGG-Sound and AVE.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

A critical analysis of self-supervision, or what we can learn from a single image

We look critically at popular self-supervision techniques for learning deep convolutional neural networks without manual labels. We show that three different and representative methods, BiGAN, RotNet and DeepCluster, can learn the first few layers of a convolutional network from a single image as well as using millions of images and manual labels, provided that strong data augmentation is used. However, for deeper layers the gap with manual supervision cannot be closed even if millions of unlabelled images are used for training. We conclude that: (1) the weights of the early layers of deep networks contain limited information about the statistics of natural images, that (2) such low-level statistics can be learned through self-supervision just as well as through strong supervision, and that (3) the low-level statistics can be captured via synthetic transformations instead of using a large image dataset.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Improving Feature Attribution through Input-specific Network Pruning

Attributing the output of a neural network to the contribution of given input elements is a way of shedding light on the black-box nature of neural networks. Due to the complexity of current network architectures, current gradient-based attribution methods provide very noisy or coarse results. We propose to prune a neural network for a given single input to keep only neurons that highly contribute to the prediction. We show that by input-specific pruning, network gradients change from reflecting local (noisy) importance information to global importance. Our proposed method is efficient and generates fine-grained attribution maps. We further provide a theoretical justification of the pruning approach relating it to perturbations and validate it through a novel experimental setup. Our method is evaluated by multiple benchmarks: sanity checks, pixel perturbation, and Remove-and-Retrain (ROAR). These benchmarks evaluate the method from different perspectives and our method performs better than other methods across all evaluations.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Self-labelling via simultaneous clustering and representation learning

Combining clustering and representation learning is one of the most promising approaches for unsupervised learning of deep neural networks. However, doing so naively leads to ill posed learning problems with degenerate solutions. In this paper, we propose a novel and principled learning formulation that addresses these issues. The method is obtained by maximizing the information between labels and input data indices. We show that this criterion extends standard crossentropy minimization to an optimal transport problem, which we solve efficiently for millions of input images and thousands of labels using a fast variant of the Sinkhorn-Knopp algorithm. The resulting method is able to self-label visual data so as to train highly competitive image representations without manual labels. Our method achieves state of the art representation learning performance for AlexNet and ResNet-50 on SVHN, CIFAR-10, CIFAR-100 and ImageNet and yields the first self-supervised AlexNet that outperforms the supervised Pascal VOC detection baseline. Code and models are available.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Semantic Image Manipulation Using Scene Graphs

Image manipulation can be considered a special case of image generation where the image to be produced is a modification of an existing image. Image generation and manipulation have been, for the most part, tasks that operate on raw pixels. However, the remarkable progress in learning rich image and object representations has opened the way for tasks such as text-to-image or layout-to-image generation that are mainly driven by semantics. In our work, we address the novel problem of image manipulation from scene graphs, in which a user can edit images by merely applying changes in the nodes or edges of a semantic graph that is generated from the image. Our goal is to encode image information in a given constellation and from there on generate new constellations, such as replacing objects or even changing relationships between objects, while respecting the semantics and style from the original image. We introduce a spatio-semantic scene graph network that does not require direct supervision for constellation changes or image edits. This makes it possible to train the system from existing real-world datasets with no additional annotation effort.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild

We propose a method to learn 3D deformable object categories from raw single-view images, without external supervision. The method is based on an autoencoder that factors each input image into depth, albedo, viewpoint and illumination. In order to disentangle these components without supervision, we use the fact that many object categories have, at least in principle, a symmetric structure. We show that reasoning about illumination allows us to exploit the underlying object symmetry even if the appearance is not symmetric due to shading. Furthermore, we model objects that are probably, but not certainly, symmetric by predicting a symmetry probability map, learned end-to-end with the other components of the model. Our experiments show that this method can recover very accurately the 3D shape of human faces, cat faces and cars from single-view images, without any supervision or a prior shape model. On benchmarks, we demonstrate superior accuracy compared to another method that uses supervision at the level of 2D image correspondences.

0 citations0 reviewsNo code linkedOpen access
preprint2019arXiv

Traffic4cast-Traffic Map Movie Forecasting -- Team MIE-Lab

The goal of the IARAI competition traffic4cast was to predict the city-wide traffic status within a 15-minute time window, based on information from the previous hour. The traffic status was given as multi-channel images (one pixel roughly corresponds to 100x100 meters), where one channel indicated the traffic volume, another one the average speed of vehicles, and a third one their rough heading. As part of our work on the competition, we evaluated many different network architectures, analyzed the statistical properties of the given data in detail, and thought about how to transform the problem to be able to take additional spatio-temporal context-information into account, such as the street network, the positions of traffic lights, or the weather. This document summarizes our efforts that led to our best submission, and gives some insights about which other approaches we evaluated, and why they did not work as well as imagined.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Deep Active Contours

We propose a method for interactive boundary extraction which combines a deep, patch-based representation with an active contour framework. We train a class-specific convolutional neural network which predicts a vector pointing from the respective point on the evolving contour towards the closest point on the boundary of the object of interest. These predictions form a vector field which is then used for evolving the contour by the Sobolev active contour framework proposed by Sundaramoorthi et al. The resulting interactive segmentation method is very efficient in terms of required computational resources and can even be trained on comparatively small graphics cards. We evaluate the potential of the proposed method on both medical and non-medical challenge data sets, such as the STACOM data set and the PASCAL VOC 2012 data set.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Deeper Depth Prediction with Fully Convolutional Residual Networks

This paper addresses the problem of estimating the depth map of a scene given a single RGB image. We propose a fully convolutional architecture, encompassing residual learning, to model the ambiguous mapping between monocular images and depth maps. In order to improve the output resolution, we present a novel way to efficiently learn feature map up-sampling within the network. For optimization, we introduce the reverse Huber loss that is particularly suited for the task at hand and driven by the value distributions commonly present in depth maps. Our model is composed of a single architecture that is trained end-to-end and does not rely on post-processing techniques, such as CRFs or other additional refinement steps. As a result, it runs in real-time on images or videos. In the evaluation, we show that the proposed model contains fewer parameters and requires fewer training data than the current state of the art, while outperforming all approaches on depth estimation. Code and models are publicly available.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Hands-Free Segmentation of Medical Volumes via Binary Inputs

We propose a novel hands-free method to interactively segment 3D medical volumes. In our scenario, a human user progressively segments an organ by answering a series of questions of the form "Is this voxel inside the object to segment?". At each iteration, the chosen question is defined as the one halving a set of candidate segmentations given the answered questions. For a quick and efficient exploration, these segmentations are sampled according to the Metropolis-Hastings algorithm. Our sampling technique relies on a combination of relaxed shape prior, learnt probability map and consistency with previous answers. We demonstrate the potential of our strategy on a prostate segmentation MRI dataset. Through the study of failure cases with synthetic examples, we demonstrate the adaptation potential of our method. We also show that our method outperforms two intuitive baselines: one based on random questions, the other one being the thresholded probability map.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Robust Optimization for Deep Regression

Convolutional Neural Networks (ConvNets) have successfully contributed to improve the accuracy of regression-based methods for computer vision tasks such as human pose estimation, landmark localization, and object detection. The network optimization has been usually performed with L2 loss and without considering the impact of outliers on the training process, where an outlier in this context is defined by a sample estimation that lies at an abnormal distance from the other training sample estimations in the objective space. In this work, we propose a regression model with ConvNets that achieves robustness to such outliers by minimizing Tukey's biweight function, an M-estimator robust to outliers, as the loss function for the ConvNet. In addition to the robust loss, we introduce a coarse-to-fine model, which processes input images of progressively higher resolutions for improving the accuracy of the regressed values. In our experiments, we demonstrate faster convergence and better generalization of our robust loss function for the tasks of human pose estimation and age estimation from face images. We also show that the combination of the robust loss function with the coarse-to-fine model produces comparable or better results than current state-of-the-art approaches in four publicly available human pose estimation datasets.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Semi-Automatic Segmentation of Autosomal Dominant Polycystic Kidneys using Random Forests

This paper presents a method for 3D segmentation of kidneys from patients with autosomal dominant polycystic kidney disease (ADPKD) and severe renal insufficiency, using computed tomography (CT) data. ADPKD severely alters the shape of the kidneys due to non-uniform formation of cysts. As a consequence, fully automatic segmentation of such kidneys is very challenging. We present a segmentation method with minimal user interaction based on a random forest classifier. One of the major novelties of the proposed approach is the usage of geodesic distance volumes as additional source of information. These volumes contain the intensity weighted distance to a manual outline of the respective kidney in only one slice (for each kidney) of the CT volume. We evaluate our method qualitatively and quantitatively on 55 CT acquisitions using ground truth annotations from clinical experts.

0 citations0 reviewsNo code linkedOpen access