no code implementations • 16 Jan 2025 • Christoph Reinders, Frederik Schubert, Bodo Rosenhahn
To address this, we introduce HydraMix, a novel architecture that generates new image compositions by mixing multiple different images from the same class.
1 code implementation • 25 Nov 2024 • Tom Wehrbein, Marco Rudolph, Bodo Rosenhahn, Bastian Wandt
Instead, we propose to additionally supervise the learned distributions by minimizing the distance to distributions encoded in heatmaps of a 2D pose detector.
1 code implementation • 21 Aug 2024 • Timo Kaiser, Vladimir Ulman, Bodo Rosenhahn
We show that CHOTA is sensitive to all tracking errors and gives a good indication of the biologically relevant capability of a method to reconstruct the full lineage of cells.
no code implementations • 29 May 2024 • Vaibhav Vavilala, Florian Kluger, Seemandhar Jain, Bodo Rosenhahn, David Forsyth
Describing a scene in terms of primitives -- geometrically simple shapes that offer a parsimonious but accurate abstraction of structure -- is an established vision problem.
1 code implementation • 10 Apr 2024 • Mathis Kruse, Marco Rudolph, Dominik Woiwode, Bodo Rosenhahn
Detecting anomalies in images has become a well-explored problem in both academia and industry.
Ranked #1 on
Anomaly Detection
on PAD Dataset
1 code implementation • 22 Mar 2024 • Timo Kaiser, Maximilian Schier, Bodo Rosenhahn
Cell tracking and segmentation assist biologists in extracting insights from large-scale microscopy time-lapse data.
no code implementations • 18 Mar 2024 • Tom Wehrbein, Bodo Rosenhahn, Iain Matthews, Carsten Stoll
To address this issue, we propose to construct dense correspondences between initial human model estimates and the corresponding images that can be used to refine the initial predictions.
no code implementations • 16 Mar 2024 • Mariia Khan, Yue Qiu, Yuren Cong, Jumana Abu-Khalaf, David Suter, Bodo Rosenhahn
The foundational Segment Anything Model (SAM) is designed for promptable multi-class multi-instance segmentation but tends to output part or sub-part masks in the "everything" mode for various real-world applications.
1 code implementation • 15 Mar 2024 • Florian Kluger, Eric Brachmann, Michael Ying Yang, Bodo Rosenhahn
A RANSAC estimator guided by a neural network fits these primitives to a depth map.
no code implementations • 5 Feb 2024 • Bodo Rosenhahn, Christoph Hirche
A Normalizing Flow computes a bijective mapping from an arbitrary distribution to a predefined (e. g. normal) distribution.
no code implementations • 5 Feb 2024 • Yannik Mahlau, Frederik Schubert, Bodo Rosenhahn
The combination of self-play and planning has achieved great successes in sequential games, for instance in Chess and Go.
1 code implementation • 26 Jan 2024 • Florian Kluger, Bodo Rosenhahn
We present a real-time method for robust estimation of multiple instances of geometric models from noisy data.
1 code implementation • 21 Dec 2023 • Thomas Norrenbrock, Marco Rudolph, Bodo Rosenhahn
Explanations in Computer Vision are often desired, but most Deep Neural Networks can only provide saliency maps with questionable faithfulness.
Ranked #1 on
Interpretable Machine Learning
on CUB-200-2011
1 code implementation • 8 Nov 2023 • Jan Thieß Brockmann, Marco Rudolph, Bodo Rosenhahn, Bastian Wandt
During the operation of industrial robots, unusual events may endanger the safety of humans and the quality of production.
Ranked #1 on
Anomaly Detection
on voraus-AD
1 code implementation • 9 Oct 2023 • Yuren Cong, Mengmeng Xu, Christian Simon, Shoufa Chen, Jiawei Ren, Yanping Xie, Juan-Manuel Perez-Rua, Bodo Rosenhahn, Tao Xiang, Sen He
In this paper, for the first time, we introduce optical flow into the attention module in the diffusion model's U-Net to address the inconsistency issue for text-to-video editing.
1 code implementation • 14 Aug 2023 • Patrick Glandorf, Timo Kaiser, Bodo Rosenhahn
Sparse neural networks are a key factor in developing resource-efficient machine learning applications.
no code implementations • 1 Aug 2023 • Andrea Avogaro, Federico Cunico, Bodo Rosenhahn, Francesco Setti
Markerless Human Pose Estimation (HPE) proved its potential to support decision making and assessment in many fields of application.
no code implementations • 5 Jun 2023 • Hendrik Hachmann, Bodo Rosenhahn
In addition, the proposed studio set is actor friendly, and produces high-quality, temporal consistent alpha and color estimations that include a superior color spill compensation.
1 code implementation • 5 Jun 2023 • Hendrik Hachmann, Bodo Rosenhahn
A maximal focus is on the accurate detection of markers and fast usage of the system.
1 code implementation • 26 Apr 2023 • Timo Kaiser, Christoph Reinders, Bodo Rosenhahn
In this paper, we propose Compensation Learning in Semantic Segmentation, a framework to identify and compensate ambiguities as well as label noise.
1 code implementation • 2 Apr 2023 • Yuren Cong, Wentong Liao, Bodo Rosenhahn, Michael Ying Yang
Learning similarity between scene graphs and images aims to estimate a similarity score given a scene graph and an image.
1 code implementation • 23 Mar 2023 • Thomas Norrenbrock, Marco Rudolph, Bodo Rosenhahn
We argue that a human can only understand the decision of a machine learning model, if the features are interpretable and only very few of them are used for a single decision.
Ranked #2 on
Interpretable Machine Learning
on CUB-200-2011
no code implementations • 4 Jan 2023 • Yuren Cong, Martin Renqiang Min, Li Erran Li, Bodo Rosenhahn, Michael Ying Yang
We further propose an attribute-centric contrastive loss to avoid overfitting to overrepresented attribute compositions.
1 code implementation • 21 Nov 2022 • Lutz M. K. Krause, Emily Manderfeld, Patricia Gnutt, Louisa Vogler, Ann Wassick, Kailey Richard, Marco Rudolph, Kelli Z. Hunsucker, Geoffrey W. Swain, Bodo Rosenhahn, Axel Rosenhahn
Biofouling is a major challenge for sustainable shipping, filter membranes, heat exchangers, and medical devices.
1 code implementation • 21 Nov 2022 • Timo Kaiser, Lukas Ehmann, Christoph Reinders, Bodo Rosenhahn
We introduce Blind Knowledge Distillation - a novel teacher-student approach for learning with noisy labels by masking the ground truth related teacher output to filter out potentially corrupted knowledge and to estimate the tipping point from generalizing to overfitting.
no code implementations • 11 Nov 2022 • Yuren Cong, Jinhui Yi, Bodo Rosenhahn, Michael Ying Yang
A semantic scene graph-to-video synthesis framework (SSGVS), based on the pre-trained VSG encoder, VQ-VAE, and auto-regressive Transformer, is proposed to synthesize a video given an initial scene image and a non-fixed number of semantic scene graphs.
1 code implementation • 14 Oct 2022 • Marco Rudolph, Tom Wehrbein, Bodo Rosenhahn, Bastian Wandt
We train a normalizing flow for density estimation as a teacher and a conventional feed-forward network as a student to trigger large distances for anomalies: The bijectivity of the normalizing flow enforces a divergence of teacher outputs for anomalies compared to normal data.
Ranked #1 on
Anomaly Detection
on MVTEC 3D-AD
(Detection AUROC metric, using extra
training data)
1 code implementation • 17 Sep 2022 • Vasileios Iosifidis, Symeon Papadopoulos, Bodo Rosenhahn, Eirini Ntoutsi
Class imbalance poses a major challenge for machine learning as most supervised learning models might exhibit bias towards the majority class and under-perform in the minority class.
no code implementations • 23 May 2022 • Frederik Schubert, Carolin Benjamins, Sebastian Döhler, Bodo Rosenhahn, Marius Lindauer
The goal of Unsupervised Reinforcement Learning (URL) is to find a reward-agnostic prior policy on a task domain, such that the sample-efficiency on supervised downstream tasks is improved.
1 code implementation • 23 Feb 2022 • Christoph Reinders, Frederik Schubert, Bodo Rosenhahn
In this work, we address the problem of learning deep neural networks on small datasets.
1 code implementation • 9 Feb 2022 • Carolin Benjamins, Theresa Eimer, Frederik Schubert, Aditya Mohan, Sebastian Döhler, André Biedenkapp, Bodo Rosenhahn, Frank Hutter, Marius Lindauer
While Reinforcement Learning ( RL) has made great strides towards solving increasingly complicated problems, many algorithms are still brittle to even slight environmental changes.
1 code implementation • 27 Jan 2022 • Yuren Cong, Michael Ying Yang, Bodo Rosenhahn
Different objects in the same scene are more or less related to each other, but only a limited number of these relationships are noteworthy.
no code implementations • CVPR 2022 • Duy M. H. Nguyen, Roberto Henschel, Bodo Rosenhahn, Daniel Sonntag, Paul Swoboda
Multi-Camera Multi-Object Tracking is currently drawing attention in the computer vision field due to its superior performance in real-world applications such as video surveillance in crowded scenes or in wide spaces.
1 code implementation • 6 Oct 2021 • Marco Rudolph, Tom Wehrbein, Bodo Rosenhahn, Bastian Wandt
In industrial manufacturing processes, errors frequently occur at unpredictable times and in unknown manifestations.
Ranked #2 on
Anomaly Detection
on Surface Defect Saliency of Magnetic Tile
(using extra training data)
1 code implementation • 5 Oct 2021 • Carolin Benjamins, Theresa Eimer, Frederik Schubert, André Biedenkapp, Bodo Rosenhahn, Frank Hutter, Marius Lindauer
While Reinforcement Learning has made great strides towards solving ever more complicated tasks, many algorithms are still brittle to even slight changes in their environment.
2 code implementations • ICCV 2021 • Andrea Hornakova, Timo Kaiser, Paul Swoboda, Michal Rolinek, Bodo Rosenhahn, Roberto Henschel
We present an efficient approximate message passing solver for the lifted disjoint paths problem (LDP), a natural but NP-hard model for multiple object tracking (MOT).
no code implementations • ICCV 2021 • Sen He, Wentong Liao, Michael Ying Yang, Yi-Zhe Song, Bodo Rosenhahn, Tao Xiang
The generated face image given a target age code is expected to be age-sensitive reflected by bio-plausible transformations of shape and texture, while being identity preserving.
1 code implementation • ICCV 2021 • Tom Wehrbein, Marco Rudolph, Bodo Rosenhahn, Bastian Wandt
3D human pose estimation from monocular images is a highly ill-posed problem due to depth ambiguities and occlusions.
Ranked #51 on
3D Human Pose Estimation
on MPI-INF-3DHP
(PCK metric)
Monocular 3D Human Pose Estimation
Multi-Hypotheses 3D Human Pose Estimation
2 code implementations • ICCV 2021 • Yuren Cong, Wentong Liao, Hanno Ackermann, Bodo Rosenhahn, Michael Ying Yang
Compared to the task of scene graph generation from images, it is more challenging because of the dynamic relationships between objects and the temporal dependencies between frames allowing for a richer semantic interpretation.
1 code implementation • 18 Jun 2021 • Maren Awiszus, Frederik Schubert, Bodo Rosenhahn
This work introduces World-GAN, the first method to perform data-driven Procedural Content Generation via Machine Learning in Minecraft from a single example.
no code implementations • 11 Jun 2021 • Frederik Schubert, Theresa Eimer, Bodo Rosenhahn, Marius Lindauer
The use of Reinforcement Learning (RL) agents in practical applications requires the consideration of suboptimal outcomes, depending on the familiarity of the agent with its environment.
Distributional Reinforcement Learning
reinforcement-learning
+2
1 code implementation • CVPR 2021 • Florian Kluger, Hanno Ackermann, Eric Brachmann, Michael Ying Yang, Bodo Rosenhahn
A RANSAC estimator guided by a neural network fits these primitives to 3D features, such as a depth map.
1 code implementation • CVPR 2022 • Kai Hu, Wentong Liao, Michael Ying Yang, Bodo Rosenhahn
Text-to-image synthesis (T2I) aims to generate photo-realistic images which are semantically consistent with the text descriptions.
1 code implementation • CVPR 2021 • Sen He, Wentong Liao, Michael Ying Yang, Yongxin Yang, Yi-Zhe Song, Bodo Rosenhahn, Tao Xiang
We argue that these are caused by the lack of context-aware object and stuff feature encoding in their generators, and location-sensitive appearance representation in their discriminators.
Ranked #1 on
Layout-to-Image Generation
on COCO-Stuff 128x128
no code implementations • 18 Mar 2021 • Hendrik Hachmann, Benjamin Krüger, Bodo Rosenhahn, Waldo Nogueira
Currently the methods used in clinics to characterize the geometry of the cochlea as well as to estimate the electrode positions are manual, error-prone and time consuming.
1 code implementation • CVPR 2021 • Bastian Wandt, Marco Rudolph, Petrissa Zell, Helge Rhodin, Bodo Rosenhahn
Human pose estimation from single images is a challenging problem in computer vision that requires large amounts of labeled training data to be solved accurately.
Ranked #3 on
3D Human Pose Estimation
on SkiPose
Monocular 3D Human Pose Estimation
Weakly-supervised 3D Human Pose Estimation
2 code implementations • 30 Oct 2020 • Hao Cheng, Wentong Liao, Xuejiao Tang, Michael Ying Yang, Monika Sester, Bodo Rosenhahn
In our framework, first, the spatial context between agents is explored by using self-attention architectures.
3 code implementations • 28 Aug 2020 • Marco Rudolph, Bastian Wandt, Bodo Rosenhahn
To achieve a high robustness and performance we exploit multiple transformations in training and evaluation.
Ranked #2 on
Anomaly Detection
on InsPLAD
2 code implementations • 4 Aug 2020 • Maren Awiszus, Frederik Schubert, Bodo Rosenhahn
In this work, we present TOAD-GAN (Token-based One-shot Arbitrary Dimension Generative Adversarial Network), a novel Procedural Content Generation (PCG) algorithm that generates token-based video game levels.
no code implementations • ECCV 2020 • Petrissa Zell, Bodo Rosenhahn, Bastian Wandt
This paper proposes a weakly-supervised learning framework for dynamics estimation from human motion.
1 code implementation • ICML 2020 • Andrea Hornakova, Roberto Henschel, Bodo Rosenhahn, Paul Swoboda
We present an extension to the disjoint paths problem in which additional \emph{lifted} edges are introduced to provide path connectivity priors.
Ranked #2 on
Multi-Object Tracking
on 2D MOT 2015
1 code implementation • 15 Jun 2020 • Hao Cheng, Wentong Liao, Michael Ying Yang, Bodo Rosenhahn, Monika Sester
Trajectory prediction is critical for applications of planning safe future movements and remains challenging even for the next few seconds in urban mixed traffic.
no code implementations • 28 May 2020 • Wentong Liao, Xiang Chen, Jingfeng Yang, Stefan Roth, Michael Goesele, Michael Ying Yang, Bodo Rosenhahn
This strengthens the local feature invariance for the resampled features and enables detecting vehicles in an arbitrary orientation.
2 code implementations • 29 Apr 2020 • Sen He, Wentong Liao, Hamed R. -Tavakoli, Michael Yang, Bodo Rosenhahn, Nicolas Pugeault
Inspired by the successes in text analysis and translation, previous work have proposed the \textit{transformer} architecture for image captioning.
1 code implementation • 5 Apr 2020 • Tongxin Hu, Vasileios Iosifidis, Wentong Liao, Hang Zhang, Michael YingYang, Eirini Ntoutsi, Bodo Rosenhahn
In this paper, we propose FairNN a neural network that performs joint feature representation and classification for fairness-aware learning.
1 code implementation • 14 Feb 2020 • Hao Cheng, Wentong Liao, Michael Ying Yang, Monika Sester, Bodo Rosenhahn
In inference time, we combine the past context and motion information of the target agent with samplings of the latent variables to predict multiple realistic trajectories in the future.
1 code implementation • ECCV 2020 • Cong Yuren, Hanno Ackermann, Wentong Liao, Michael Ying Yang, Bodo Rosenhahn
Detected objects, their labels and the discovered relations can be used to construct a scene graph which provides an abstract semantic interpretation of an image.
Ranked #8 on
Scene Graph Generation
on Visual Genome
3 code implementations • CVPR 2020 • Florian Kluger, Eric Brachmann, Hanno Ackermann, Carsten Rother, Michael Ying Yang, Bodo Rosenhahn
We present a robust estimator for fitting multiple parametric models of the same form to noisy measurements.
no code implementations • 25 Nov 2019 • Christoph Reinders, Bodo Rosenhahn
We present Neural Random Forest Imitation - a novel approach for transforming random forests into neural networks.
no code implementations • 26 Aug 2019 • Maren Awiszus, Hanno Ackermann, Bodo Rosenhahn
We use face images as our example of choice.
no code implementations • 7 Aug 2019 • Marco Rudolph, Bastian Wandt, Bodo Rosenhahn
In this paper we propose Structuring AutoEncoders (SAE).
1 code implementation • 23 Jul 2019 • Florian Kluger, Hanno Ackermann, Michael Ying Yang, Bodo Rosenhahn
The horizon line is an important geometric feature for many image processing and scene understanding tasks in computer vision.
Ranked #1 on
Horizon Line Estimation
on KITTI Horizon
no code implementations • The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops 2019 • Roberto Henschel, Yunzhe Zou, Bodo Rosenhahn
We evaluate our framework on the MOT16/17 benchmark.
Ranked #39 on
Multi-Object Tracking
on MOT17
(MOTA metric)
no code implementations • 3 Apr 2019 • Wentong Liao, Cuiling Lan, Wen-Jun Zeng, Michael Ying Yang, Bodo Rosenhahn
We further explore more powerful representations by integrating language prior with the visual context in the transformation for the scene graph generation.
1 code implementation • CVPR 2019 • Bastian Wandt, Bodo Rosenhahn
This efficiently avoids a simple memorization of the training data and allows for a weakly supervised training.
Ranked #24 on
Weakly-supervised 3D Human Pose Estimation
on Human3.6M
no code implementations • 26 Oct 2018 • Michael Ying Yang, Wentong Liao, Chun Yang, Yanpeng Cao, Bodo Rosenhahn
The experimental results show that the proposed approach outperforms the state-of-the-art methods and effective in recognizing complex security events.
no code implementations • ECCV 2018 • Timo von Marcard, Roberto Henschel, Michael J. Black, Bodo Rosenhahn, Gerard Pons-Moll
In this work, we propose a method that combines a single hand-held camera and a set of Inertial Measurement Units (IMUs) attached at the body limbs to estimate accurate 3D poses in the wild.
no code implementations • 2 May 2018 • Maren Awiszus, Bodo Rosenhahn
In this work we present a modified neural network model which is capable to simulate Markov Chains.
no code implementations • 9 Feb 2018 • Michael Ying Yang, Wentong Liao, Yanpeng Cao, Bodo Rosenhahn
In our framework, three levels of video events are connected by Hierarchical Dirichlet Process (HDP) model: low-level visual features, simple atomic activities, and multi-agent interactions.
no code implementations • 9 Feb 2018 • Oliver Mueller, Michael Ying Yang, Bodo Rosenhahn
We propose to avoid dependence on a proposal distribution by introducing a slice sampling based PBP algorithm.
no code implementations • 9 Feb 2018 • Michael Ying Yang, Matthias Reso, Jun Tang, Wentong Liao, Bodo Rosenhahn
Therefore, we formulate a graphical model to select a proposal stream for each object in which the pairwise potentials consist of the appearance dissimilarity between different streams in the same video and also the similarity between the streams in different videos.
1 code implementation • 9 Feb 2018 • Wentong Liao, Michael Ying Yang, Ni Zhan, Bodo Rosenhahn
Moreover, we trained the model jointly on six different datasets, which differs from common practice - one model is just trained on one dataset and tested also on the same one.
no code implementations • 22 Jan 2018 • Michael Ying Yang, Wentong Liao, Xinbo Li, Bodo Rosenhahn
Also, the focal loss function is used to substitute for conventional cross entropy loss function in both of the region proposed network and the final classifier.
no code implementations • 16 Nov 2017 • Wentong Liao, Lin Shuai, Bodo Rosenhahn, Michael Ying Yang
Most of the existing works treat this task as a pure visual classification task: each type of relationship or phrase is classified as a relation category based on the extracted visual features.
no code implementations • 18 Sep 2017 • Christoph Reinders, Hanno Ackermann, Michael Ying Yang, Bodo Rosenhahn
These algorithms are usually trained on large datasets consisting of thousands or millions of labeled training examples.
2 code implementations • 8 Jul 2017 • Florian Kluger, Hanno Ackermann, Michael Ying Yang, Bodo Rosenhahn
We present a novel approach for vanishing point detection from uncalibrated monocular images.
Ranked #3 on
Horizon Line Estimation
on York Urban Dataset
no code implementations • 8 Jun 2017 • Dario Augusto Borges Oliveira, Laura Leal-Taixe, Raul Queiroz Feitosa, Bodo Rosenhahn
Further visual results also show the potential of our approach for identifying vascular networks topologies.
no code implementations • 23 May 2017 • Roberto Henschel, Laura Leal-Taixé, Daniel Cremers, Bodo Rosenhahn
In order to track all persons in a scene, the tracking-by-detection paradigm has proven to be a very effective approach.
Ranked #22 on
Multi-Object Tracking
on MOT16
no code implementations • 23 Mar 2017 • Timo von Marcard, Bodo Rosenhahn, Michael J. Black, Gerard Pons-Moll
We address the problem of making human motion capture in the wild more practical by using a small set of inertial sensors attached to the body.
no code implementations • 1 Feb 2017 • Bastian Wandt, Hanno Ackermann, Bodo Rosenhahn
This paper deals with motion capture of kinematic chains (e. g. human skeletons) from monocular image sequences taken by uncalibrated cameras.
no code implementations • 24 Jan 2017 • Michael Ying Yang, Hanno Ackermann, Weiyao Lin, Sitong Feng, Bodo Rosenhahn
In this paper, we propose a new framework for segmenting feature-based moving objects under affine subspace model.
no code implementations • 19 Sep 2016 • Michael Ying Yang, Wentong Liao, Hanno Ackermann, Bodo Rosenhahn
In contrast to previous methods for extracting support relations, the proposed approach generates more accurate results, and does not require a pixel-wise semantic labeling of the scene.
no code implementations • 16 Sep 2016 • Hanno Ackermann, Michael Ying Yang, Bodo Rosenhahn
If these unknown subspaces are well-separated this algorithm is guaranteed to succeed.
no code implementations • 25 Jul 2016 • Roberto Henschel, Laura Leal-Taixé, Bodo Rosenhahn, Konrad Schindler
We present a novel formulation of the multiple object tracking problem which integrates low and mid-level features.
no code implementations • CVPR 2016 • Eduardo Perez-Pellitero, Jordi Salvador, Javier Ruiz-Hidalgo, Bodo Rosenhahn
The main challenge in Super Resolution (SR) is to discover the mapping between the low- and high-resolution manifolds of image patches, a complex ill-posed problem which has recently been addressed through piecewise linear regression with promising results.
no code implementations • CVPR 2016 • Andrey Bushnevskiy, Lorenzo Sorgi, Bodo Rosenhahn
Multicamera rigs are used in a large number of 3D Vision applications, such as 3D modeling, motion capture or telepresence and a robust calibration is of utmost importance in order to achieve a high accuracy results.
no code implementations • 6 Aug 2015 • Saif Dawood Salman Al-Shaikhli, Michael Ying Yang, Bodo Rosenhahn
A sparse representation of both global (region-based) and local (voxel-wise) image information is embedded in a level set formulation to innovate a new cost function.
no code implementations • CVPR 2015 • Alina Kuznetsova, Sung Ju Hwang, Bodo Rosenhahn, Leonid Sigal
By incrementally detecting object instances in video and adding confident detections into the model, we are able to dynamically adjust the complexity of the detector over time by instantiating new prototypes to span all domains the model has seen.
no code implementations • CVPR 2014 • Laura Leal-Taixe, Michele Fenzi, Alina Kuznetsova, Bodo Rosenhahn, Silvio Savarese
We present a novel method for multiple people tracking that leverages a generalized model for capturing interactions among individuals.
no code implementations • CVPR 2014 • Gerard Pons-Moll, David J. Fleet, Bodo Rosenhahn
We advocate the inference of qualitative information about 3D human pose, called posebits, from images.
no code implementations • CVPR 2013 • Michele Fenzi, Laura Leal-Taixe, Bodo Rosenhahn, Jorn Ostermann
In this paper, we propose a method for learning a class representation that can return a continuous value for the pose of an unknown class instance using only 2D data and weak 3D labelling information.