Daniel Kienzle
YOU?
Author Swipe
View article: MMMS: Multi-Modal Multi-Surface Interactive Segmentation
MMMS: Multi-Modal Multi-Surface Interactive Segmentation Open
In this paper, we present a method to interactively create segmentation masks on the basis of user clicks. We pay particular attention to the segmentation of multiple surfaces that are simultaneously present in the same image. Since these …
View article: Towards Ball Spin and Trajectory Analysis in Table Tennis Broadcast Videos via Physically Grounded Synthetic-to-Real Transfer
Towards Ball Spin and Trajectory Analysis in Table Tennis Broadcast Videos via Physically Grounded Synthetic-to-Real Transfer Open
Analyzing a player's technique in table tennis requires knowledge of the ball's 3D trajectory and spin. While, the spin is not directly observable in standard broadcasting videos, we show that it can be inferred from the ball's trajectory …
View article: Efficient 2D to Full 3D Human Pose Uplifting including Joint Rotations
Efficient 2D to Full 3D Human Pose Uplifting including Joint Rotations Open
In sports analytics, accurately capturing both the 3D locations and rotations of body joints is essential for understanding an athlete's biomechanics. While Human Mesh Recovery (HMR) models can estimate joint rotations, they often exhibit …
View article: SkipClick: Combining Quick Responses and Low-Level Features for Interactive Segmentation in Winter Sports Contexts
SkipClick: Combining Quick Responses and Low-Level Features for Interactive Segmentation in Winter Sports Contexts Open
In this paper, we present a novel architecture for interactive segmentation in winter sports contexts. The field of interactive segmentation deals with the prediction of high-quality segmentation masks by informing the network about the ob…
View article: Leveraging Anthropometric Measurements to Improve Human Mesh Estimation and Ensure Consistent Body Shapes
Leveraging Anthropometric Measurements to Improve Human Mesh Estimation and Ensure Consistent Body Shapes Open
The basic body shape (i.e., the body shape in T-pose) of a person does not change within a single video. However, most SOTA human mesh estimation (HME) models output a slightly different, thus inconsistent basic body shape for each video f…
View article: WSESeg: Introducing a Dataset for the Segmentation of Winter Sports Equipment with a Baseline for Interactive Segmentation
WSESeg: Introducing a Dataset for the Segmentation of Winter Sports Equipment with a Baseline for Interactive Segmentation Open
In this paper we introduce a new dataset containing instance segmentation masks for ten different categories of winter sports equipment, called WSESeg (Winter Sports Equipment Segmentation). Furthermore, we carry out interactive segmentati…
View article: A Fair Ranking and New Model for Panoptic Scene Graph Generation
A Fair Ranking and New Model for Panoptic Scene Graph Generation Open
In panoptic scene graph generation (PSGG), models retrieve interactions between objects in an image which are grounded by panoptic segmentation masks. Previous evaluations on panoptic scene graphs have been subject to an erroneous evaluati…
View article: The STOIC2021 COVID-19 AI challenge: Applying reusable training methodologies to private data
The STOIC2021 COVID-19 AI challenge: Applying reusable training methodologies to private data Open
Challenges drive the state-of-the-art of automated medical image analysis. The quantity of public training data that they provide can limit the performance of their solutions. Public access to the training methodology for these solutions r…
View article: Segformer++: Efficient Token-Merging Strategies for High-Resolution Semantic Segmentation
Segformer++: Efficient Token-Merging Strategies for High-Resolution Semantic Segmentation Open
Utilizing transformer architectures for semantic segmentation of high-resolution images is hindered by the attention's quadratic computational complexity in the number of tokens. A solution to this challenge involves decreasing the number …
View article: Towards Learning Monocular 3D Object Localization From 2D Labels using the Physical Laws of Motion
Towards Learning Monocular 3D Object Localization From 2D Labels using the Physical Laws of Motion Open
We present a novel method for precise 3D object localization in single images from a single calibrated camera using only 2D labels. No expensive 3D labels are needed. Thus, instead of using 3D labels, our model is trained with easy-to-anno…
View article: Haystack: A Panoptic Scene Graph Dataset to Evaluate Rare Predicate Classes
Haystack: A Panoptic Scene Graph Dataset to Evaluate Rare Predicate Classes Open
Current scene graph datasets suffer from strong long-tail distributions of their predicate classes. Due to a very low number of some predicate classes in the test sets, no reliable metrics can be retrieved for the rarest classes. We constr…
View article: Haystack: A Panoptic Scene Graph Dataset to Evaluate Rare Predicate Classes
Haystack: A Panoptic Scene Graph Dataset to Evaluate Rare Predicate Classes Open
Current scene graph datasets suffer from strong long-tail distributions of their predicate classes. Due to a very low number of some predicate classes in the test sets, no reliable metrics can be retrieved for the rarest classes. We constr…
View article: The STOIC2021 COVID-19 AI challenge: applying reusable training methodologies to private data
The STOIC2021 COVID-19 AI challenge: applying reusable training methodologies to private data Open
Challenges drive the state-of-the-art of automated medical image analysis. The quantity of public training data that they provide can limit the performance of their solutions. Public access to the training methodology for these solutions r…
View article: Detecting Arbitrary Keypoints on Limbs and Skis with Sparse Partly Correct Segmentation Masks
Detecting Arbitrary Keypoints on Limbs and Skis with Sparse Partly Correct Segmentation Masks Open
Analyses based on the body posture are crucial for top-class athletes in many sports disciplines. If at all, coaches label only the most important keypoints, since manual annotations are very costly. This paper proposes a method to detect …
View article: COVID Detection and Severity Prediction with 3D-ConvNeXt and Custom Pretrainings
COVID Detection and Severity Prediction with 3D-ConvNeXt and Custom Pretrainings Open
Since COVID strongly affects the respiratory system, lung CT-scans can be used for the analysis of a patients health. We introduce a neural network for the prediction of the severity of lung damage and the detection of a COVID-infection us…
View article: Recognition of Freely Selected Keypoints on Human Limbs
Recognition of Freely Selected Keypoints on Human Limbs Open
Nearly all Human Pose Estimation (HPE) datasets consist of a fixed set of keypoints. Standard HPE models trained on such datasets can only detect these keypoints. If more points are desired, they have to be manually annotated and the model…