Bastian Wandt
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Bastian Wandt has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Helge Rhodin
Bielefeld University, Bielefeld, Germany and Max Planck Institute for Informatics, Saarbruecken, Germany
3 shared publications - Bodo Rosenhahn
Leibniz University Hannover
2 shared publications - Jan Thieß Brockmann
Voraus robotik GmbH, Hanover, Germany
2 shared publications - Marco Rudolph
Institute for Information Processing, L3S, Leibniz University Hannover, Hannover, Germany
2 shared publications - Chunjin Song
University of British Columbia
1 shared publication - Chunjin Song
University of British Columbia
1 shared publication
Latest Publications
- Open Access
The Effective Inertia of the Lower Limb During Locomotion
Accurate modelling of inertial properties of human lower limbs is of great interest to many tasks, from gait analysis in biomechanics to motion tracking and control in computer animation. Previous work typically simplified the human musculoskeletal structure as a chain of rigid capsules, with muscle mass lumped with body segments. Such simplifications lead to errors in the inertia matrix, and the error propagates to torque and pose estimates. In this study, we developed a data-driven model to represent the joint-space inertia of the lower-body of a human in motion. The model does not make any assumptions other than that the estimated inertia matrices must be symmetric and positive definite. We show that a joint-space inertia matrix, estimated from synchronized motion and ground force data following foot strikes, reveals inertial coupling, and that estimated inertia matrices are bilaterally symmetric and motion-type dependent. These are properties which a rigid, mass-lumped inertia matrix fails to entail. Moreover, we show that the data-driven model fits to data better than the articulated rigid body inertia model, and that when used for reconstructing lower body kinematics estimated inertia yields more accurate and stable motion.
Open AccessThe Effective Inertia of the Lower Limb During Locomotion
Accurate modelling of inertial properties of human lower limbs is of great interest to many tasks, from gait analysis in biomechanics to motion tracking and control in computer animation. Previous work typically simplified the human musculoskeletal structure as a chain of rigid capsules, with muscle mass lumped with body segments. Such simplifications lead to errors in the inertia matrix, and the error propagates to torque and pose estimates. In this study, we developed a data-driven model to represent the joint-space inertia of the lower-body of a human in motion. The model does not make any assumptions other than that the estimated inertia matrices must be symmetric and positive definite. We show that a joint-space inertia matrix, estimated from synchronized motion and ground force data following foot strikes, reveals inertial coupling, and that estimated inertia matrices are bilaterally symmetric and motion-type dependent. These are properties which a rigid, mass-lumped inertia matrix fails to entail. Moreover, we show that the data-driven model fits to data better than the articulated rigid body inertia model, and that when used for reconstructing lower body kinematics estimated inertia yields more accurate and stable motion.
- Open Access
- 28 August 2025
- Published by Wiley in Computer Graphics Forum
- Vol. 44 (5)
Representing Animatable Avatar via Factorized Neural Fields
For reconstructing high‐fidelity human 3D models from monocular videos, it is crucial to maintain consistent large‐scale body shapes along with finely matched subtle wrinkles. This paper explores how per‐frame rendering results can be factorized into a pose‐independent component and a corresponding pose‐dependent counterpart to facilitate frame consistency at multiple scales. Pose adaptive texture features are further improved by restricting the frequency bands of these two components. Pose‐independent outputs are expected to be low‐frequency, while high‐frequency information is linked to pose‐dependent factors. We implement this with a dual‐branch network. The first branch takes coordinates in the canonical space as input, while the second one additionally considers features outputted by the first branch and pose information of each frame. A final network integrates the information predicted by both branches and utilizes volume rendering to generate photo‐realistic 3D human images. Through experiments, we demonstrate that our method consistently surpasses all state‐of‐the‐art methods in preserving high‐frequency details and ensuring consistent body contours. Our code is accessible at https://github.com/ChunjinSong/facavatar.
- 28 August 2025
- Published by Wiley in Computer Graphics Forum
- Vol. 44 (5)
Open AccessRepresenting Animatable Avatar via Factorized Neural Fields
For reconstructing high‐fidelity human 3D models from monocular videos, it is crucial to maintain consistent large‐scale body shapes along with finely matched subtle wrinkles. This paper explores how per‐frame rendering results can be factorized into a pose‐independent component and a corresponding pose‐dependent counterpart to facilitate frame consistency at multiple scales. Pose adaptive texture features are further improved by restricting the frequency bands of these two components. Pose‐independent outputs are expected to be low‐frequency, while high‐frequency information is linked to pose‐dependent factors. We implement this with a dual‐branch network. The first branch takes coordinates in the canonical space as input, while the second one additionally considers features outputted by the first branch and pose information of each frame. A final network integrates the information predicted by both branches and utilizes volume rendering to generate photo‐realistic 3D human images. Through experiments, we demonstrate that our method consistently surpasses all state‐of‐the‐art methods in preserving high‐frequency details and ensuring consistent body contours. Our code is accessible at https://github.com/ChunjinSong/facavatar.
- 16 June 2025
- Published by Springer Nature
- p. 276-291
Continuous Normalizing Flows for Uncertainty-Aware Human Pose Estimation
- 16 June 2025
- Published by Springer Nature
- p. 276-291
Continuous Normalizing Flows for Uncertainty-Aware Human Pose Estimation
- Open Access
Temporally-consistent 3D Reconstruction of Birds
This paper deals with 3D reconstruction of seabirds which recently came into focus of environmental scientists as valuable bio-indicators for environmental change. Such 3D information is beneficial for analyzing the bird's behavior and physiological shape, for example by tracking motion, shape, and appearance changes. From a computer vision perspective birds are especially challenging due to their rapid and oftentimes non-rigid motions. We propose an approach to reconstruct the 3D pose and shape from monocular videos of a specific breed of seabird - the common murre. Our approach comprises a full pipeline of detection, tracking, segmentation, and temporally consistent 3D reconstruction. Additionally, we propose a temporal loss that extends current single-image 3D bird pose estimators to the temporal domain. Moreover, we provide a real-world dataset of 10000 frames of video observations on average capture nine birds simultaneously, comprising a large variety of motions and interactions, including a smaller test set with bird-specific keypoint labels. Using our temporal optimization, we achieve state-of-the-art performance for the challenging sequences in our dataset.
Open AccessTemporally-consistent 3D Reconstruction of Birds
This paper deals with 3D reconstruction of seabirds which recently came into focus of environmental scientists as valuable bio-indicators for environmental change. Such 3D information is beneficial for analyzing the bird's behavior and physiological shape, for example by tracking motion, shape, and appearance changes. From a computer vision perspective birds are especially challenging due to their rapid and oftentimes non-rigid motions. We propose an approach to reconstruct the 3D pose and shape from monocular videos of a specific breed of seabird - the common murre. Our approach comprises a full pipeline of detection, tracking, segmentation, and temporally consistent 3D reconstruction. Additionally, we propose a temporal loss that extends current single-image 3D bird pose estimators to the temporal domain. Moreover, we provide a real-world dataset of 10000 frames of video observations on average capture nine birds simultaneously, comprising a large variety of motions and interactions, including a smaller test set with bird-specific keypoint labels. Using our temporal optimization, we achieve state-of-the-art performance for the challenging sequences in our dataset.
- 13 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 40 (15523098) , 438-451
The voraus-AD Dataset for Anomaly Detection in Robot Applications
During the operation of industrial robots, unusual events may endanger the safety of humans and the quality of production. When collecting data to detect such cases, it is not ensured that data from all potentially occurring errors is included as unforeseeable events may happen over time. Therefore, anomaly detection (AD) delivers a practical solution, using only normal data to learn to detect unusual events. We introduce a dataset that allows training and benchmarking of anomaly detection methods for robotic applications based on machine data which will be made publicly available to the research community. As a typical robot task the dataset includes a pick-and-place application which involves movement, actions of the end effector, and interactions with the objects of the environment. Since several of the contained anomalies are not task-specific but general, evaluations on our dataset are transferable to other robotics applications as well. In addition, we present multivariate time-series flow (MVT-Flow) as a new baseline method for anomaly detection: It relies on deep-learning-based density estimation with normalizing flows, tailored to the data domain by taking its structure into account for the architecture. Our evaluation shows that MVT-Flow outperforms baselines from previous work by a large margin of 6.2% in area under receiving operator characteristic.
- 13 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 40 (15523098) , 438-451
The voraus-AD Dataset for Anomaly Detection in Robot Applications
During the operation of industrial robots, unusual events may endanger the safety of humans and the quality of production. When collecting data to detect such cases, it is not ensured that data from all potentially occurring errors is included as unforeseeable events may happen over time. Therefore, anomaly detection (AD) delivers a practical solution, using only normal data to learn to detect unusual events. We introduce a dataset that allows training and benchmarking of anomaly detection methods for robotic applications based on machine data which will be made publicly available to the research community. As a typical robot task the dataset includes a pick-and-place application which involves movement, actions of the end effector, and interactions with the objects of the environment. Since several of the contained anomalies are not task-specific but general, evaluations on our dataset are transferable to other robotics applications as well. In addition, we present multivariate time-series flow (MVT-Flow) as a new baseline method for anomaly detection: It relies on deep-learning-based density estimation with normalizing flows, tailored to the data domain by taking its structure into account for the architecture. Our evaluation shows that MVT-Flow outperforms baselines from previous work by a large margin of 6.2% in area under receiving operator characteristic.
- Open Access
The voraus-AD Dataset for Anomaly Detection in Robot Applications
During the operation of industrial robots, unusual events may endanger the safety of humans and the quality of production. When collecting data to detect such cases, it is not ensured that data from all potentially occurring errors is included as unforeseeable events may happen over time. Therefore, anomaly detection (AD) delivers a practical solution, using only normal data to learn to detect unusual events. We introduce a dataset that allows training and benchmarking of anomaly detection methods for robotic applications based on machine data which will be made publicly available to the research community. As a typical robot task the dataset includes a pick-and-place application which involves movement, actions of the end effector and interactions with the objects of the environment. Since several of the contained anomalies are not task-specific but general, evaluations on our dataset are transferable to other robotics applications as well. Additionally, we present MVT-Flow (multivariate time-series flow) as a new baseline method for anomaly detection: It relies on deep-learning-based density estimation with normalizing flows, tailored to the data domain by taking its structure into account for the architecture. Our evaluation shows that MVT-Flow outperforms baselines from previous work by a large margin of 6.2% in area under ROC.
Open AccessThe voraus-AD Dataset for Anomaly Detection in Robot Applications
During the operation of industrial robots, unusual events may endanger the safety of humans and the quality of production. When collecting data to detect such cases, it is not ensured that data from all potentially occurring errors is included as unforeseeable events may happen over time. Therefore, anomaly detection (AD) delivers a practical solution, using only normal data to learn to detect unusual events. We introduce a dataset that allows training and benchmarking of anomaly detection methods for robotic applications based on machine data which will be made publicly available to the research community. As a typical robot task the dataset includes a pick-and-place application which involves movement, actions of the end effector and interactions with the objects of the environment. Since several of the contained anomalies are not task-specific but general, evaluations on our dataset are transferable to other robotics applications as well. Additionally, we present MVT-Flow (multivariate time-series flow) as a new baseline method for anomaly detection: It relies on deep-learning-based density estimation with normalizing flows, tailored to the data domain by taking its structure into account for the architecture. Our evaluation shows that MVT-Flow outperforms baselines from previous work by a large margin of 6.2% in area under ROC.
- Open Access
Pose Modulated Avatars from Video
It is now possible to reconstruct dynamic human motion and shape from a sparse set of cameras using Neural Radiance Fields (NeRF) driven by an underlying skeleton. However, a challenge remains to model the deformation of cloth and skin in relation to skeleton pose. Unlike existing avatar models that are learned implicitly or rely on a proxy surface, our approach is motivated by the observation that different poses necessitate unique frequency assignments. Neglecting this distinction yields noisy artifacts in smooth areas or blurs fine-grained texture and shape details in sharp regions. We develop a two-branch neural network that is adaptive and explicit in the frequency domain. The first branch is a graph neural network that models correlations among body parts locally, taking skeleton pose as input. The second branch combines these correlation features to a set of global frequencies and then modulates the feature encoding. Our experiments demonstrate that our network outperforms state-of-the-art methods in terms of preserving details and generalization capabilities.
Open AccessPose Modulated Avatars from Video
It is now possible to reconstruct dynamic human motion and shape from a sparse set of cameras using Neural Radiance Fields (NeRF) driven by an underlying skeleton. However, a challenge remains to model the deformation of cloth and skin in relation to skeleton pose. Unlike existing avatar models that are learned implicitly or rely on a proxy surface, our approach is motivated by the observation that different poses necessitate unique frequency assignments. Neglecting this distinction yields noisy artifacts in smooth areas or blurs fine-grained texture and shape details in sharp regions. We develop a two-branch neural network that is adaptive and explicit in the frequency domain. The first branch is a graph neural network that models correlations among body parts locally, taking skeleton pose as input. The second branch combines these correlation features to a set of global frequencies and then modulates the feature encoding. Our experiments demonstrate that our network outperforms state-of-the-art methods in terms of preserving details and generalization capabilities.