1 code implementation • 12 Apr 2024 • Nadezda Kirillova, M. Jehanzeb Mirza, Horst Possegger, Horst Bischof
To address these limitations, we propose a pipeline for physic-based volumetric fog simulation in arbitrary real-world MOT dataset utilizing frame-by-frame monocular depth estimation and a fog formation optical model.
no code implementations • 21 Mar 2024 • Jakub Micorek, Horst Possegger, Dominik Narnhofer, Horst Bischof, Mateusz Kozinski
We propose a novel approach to video anomaly detection: we treat feature vectors extracted from videos as realizations of a random variable with a fixed distribution and model this distribution with a neural network.
no code implementations • 16 Jan 2024 • Abhiram Kolli, Filippo Casamassima, Horst Possegger, Horst Bischof
Using neural networks for localization of key fob within and surrounding a car as a security feature for keyless entry is fast emerging.
1 code implementation • ICCV 2023 • David Schinagl, Georg Krispel, Christian Fruhwirth-Reisinger, Horst Possegger, Horst Bischof
Widely-used LiDAR-based 3D object detectors often neglect fundamental geometric information readily available from the object proposals in their confidence estimation.
1 code implementation • 13 Sep 2023 • M. Jehanzeb Mirza, Leonid Karlinsky, Wei Lin, Horst Possegger, Rogerio Feris, Horst Bischof
Vision and Language Models (VLMs), such as CLIP, have enabled visual recognition of a potentially unlimited set of categories described by text prompts.
1 code implementation • 30 May 2023 • Stefan Leitner, M. Jehanzeb Mirza, Wei Lin, Jakub Micorek, Marc Masana, Mateusz Kozinski, Horst Possegger, Horst Bischof
We propose to store these affine parameters as a memory bank for each weather condition and plug-in their weather-specific parameters during driving (i. e. test time) when the respective weather conditions are encountered.
no code implementations • 27 Mar 2023 • Kunyang Sun, Wei Lin, Haoqin Shi, Zhengming Zhang, Yongming Huang, Horst Bischof
This results in an imbalance of the adversarial training between the domain discriminator and the feature extractor.
1 code implementation • ICCV 2023 • Wei Lin, Leonid Karlinsky, Nina Shvetsova, Horst Possegger, Mateusz Kozinski, Rameswar Panda, Rogerio Feris, Hilde Kuehne, Horst Bischof
We adapt a VL model for zero-shot and few-shot action recognition using a collection of unlabeled videos and an unpaired action dictionary.
Ranked #3 on Zero-Shot Action Recognition on Kinetics
no code implementations • 9 Mar 2023 • Wei Lin, Anna Kukleva, Horst Possegger, Hilde Kuehne, Horst Bischof
Temporal action segmentation in untrimmed videos has gained increased attention recently.
no code implementations • 14 Dec 2022 • Georg Krispel, David Schinagl, Christian Fruhwirth-Reisinger, Horst Possegger, Horst Bischof
The sensing process of large-scale LiDAR point clouds inevitably causes large blind spots, i. e. regions not visible to the sensor.
no code implementations • 6 Dec 2022 • Bisheng Wang, Horst Possegger, Horst Bischof, Guo Cao
Existing Multiple Object Tracking (MOT) methods design complex architectures for better tracking performance.
1 code implementation • CVPR 2023 • Wei Lin, Muhammad Jehanzeb Mirza, Mateusz Kozinski, Horst Possegger, Hilde Kuehne, Horst Bischof
Our proposed method demonstrates a substantial performance gain over existing test-time adaptation approaches in both evaluations of a single distribution shift and the challenging case of random distribution shifts.
1 code implementation • CVPR 2023 • Muhammad Jehanzeb Mirza, Pol Jané Soneira, Wei Lin, Mateusz Kozinski, Horst Possegger, Horst Bischof
Test-Time-Training (TTT) is an approach to cope with out-of-distribution (OOD) data by adapting a trained model to distribution shifts occurring at test-time.
1 code implementation • ICCV 2023 • M. Jehanzeb Mirza, Inkyu Shin, Wei Lin, Andreas Schriebl, Kunyang Sun, Jaesung Choe, Horst Possegger, Mateusz Kozinski, In So Kweon, Kun-Jin Yoon, Horst Bischof
Our MATE is the first Test-Time-Training (TTT) method designed for 3D data, which makes deep networks trained for point cloud classification robust to distribution shifts occurring in test data.
no code implementations • 10 Nov 2022 • Abhiram Kolli, Muhammad Jehanzeb Mirza, Horst Possegger, Horst Bischof
Keyless entry systems in cars are adopting neural networks for localizing its operators.
1 code implementation • 14 Oct 2022 • Dušan Malić, Christian Fruhwirth-Reisinger, Horst Possegger, Horst Bischof
State-of-the-art unsupervised domain adaptation approaches outsource methods to overcome the object size bias.
1 code implementation • 19 Apr 2022 • M. Jehanzeb Mirza, Marc Masana, Horst Possegger, Horst Bischof
This catastrophic forgetting is typically addressed via incremental learning approaches which usually re-train the model by either keeping a memory bank of training samples or keeping a copy of the entire model or model parameters for each scenario.
1 code implementation • CVPR 2022 • David Schinagl, Georg Krispel, Horst Possegger, Peter M. Roth, Horst Bischof
These maps indicate the importance of each 3D point in predicting the specific objects.
1 code implementation • 30 Mar 2022 • Wei Lin, Anna Kukleva, Kunyang Sun, Horst Possegger, Hilde Kuehne, Horst Bischof
To address these challenges, we propose Cycle Domain Adaptation (CycDA), a cycle-based approach for unsupervised image-to-video domain adaptation by leveraging the joint spatial information in images and videos on the one hand and, on the other hand, training an independent spatio-temporal model to bridge the modality gap.
no code implementations • 20 Mar 2022 • Niloofar Azizi, Horst Possegger, Emanuele Rodolà, Horst Bischof
In particular, this allows us to directly and explicitly encode the transformation between joints, resulting in a significantly more compact representation.
Ranked #58 on 3D Human Pose Estimation on Human3.6M
1 code implementation • CVPR 2022 • M. Jehanzeb Mirza, Jakub Micorek, Horst Possegger, Horst Bischof
This can be a hurdle in fields which require continuous dynamic adaptation or suffer from scarcity of data, e. g. autonomous driving in challenging weather conditions.
1 code implementation • 26 Nov 2021 • Franz Thaler, Christian Payer, Horst Bischof, Darko Stern
Even though many semantic segmentation methods exist that are able to perform well on many medical datasets, often, they are not designed for direct use in clinical practice.
no code implementations • 18 Oct 2021 • Christian Fruhwirth-Reisinger, Michael Opitz, Horst Possegger, Horst Bischof
In the field of autonomous driving, self-training is widely applied to mitigate distribution shifts in LiDAR-based 3D object detectors.
no code implementations • 16 Jun 2021 • Muhammad Jehanzeb Mirza, Cornelius Buerkle, Julio Jarquin, Michael Opitz, Fabian Oboril, Kay-Ulrich Scholl, Horst Bischof
State-of-the-art object detection systems for autonomous driving achieve promising results in clear weather conditions.
no code implementations • 18 Dec 2019 • Georg Krispel, Michael Opitz, Georg Waltner, Horst Possegger, Horst Bischof
We introduce a simple yet effective fusion method of LiDAR and RGB data to segment LiDAR point clouds.
3 code implementations • 2 Aug 2019 • Christian Payer, Darko Štern, Horst Bischof, Martin Urschler
In many medical image analysis applications, often only a limited amount of training data is available, which makes training of convolutional neural networks (CNNs) challenging.
1 code implementation • 23 Nov 2018 • Georg Poier, Michael Opitz, David Schinagl, Horst Bischof
In this work, we remove this requirement by learning to map from the features of real data to the features of synthetic data mainly using a large amount of synthetic and unlabeled real data.
no code implementations • ECCV 2018 • Thomas Holzmann, Michael Maurer, Friedrich Fraundorfer, Horst Bischof
We propose a method for urban 3D reconstruction, which incorporates semantic information and plane priors within the reconstruction process in order to generate visually appealing 3D models.
no code implementations • 6 Jun 2018 • Christian Payer, Darko Štern, Thomas Neff, Horst Bischof, Martin Urschler
Furthermore, we train the network with a novel embedding loss based on cosine similarities, such that the network predicts unique embeddings for every instance throughout videos.
no code implementations • 9 May 2018 • Georg Waltner, Michael Maurer, Thomas Holzmann, Patrick Ruprecht, Michael Opitz, Horst Possegger, Friedrich Fraundorfer, Horst Bischof
Furthermore due to the design of the network, at test time only the 2D camera images are required for classification which enables the usage in portable computer vision systems.
2 code implementations • CVPR 2018 • Georg Poier, David Schinagl, Horst Bischof
To exploit this observation, we train a model that -- given input from one view -- estimates a latent representation, which is trained to be predictive for the appearance of the object when captured from another viewpoint.
no code implementations • 22 Mar 2018 • Christian Mostegel, Friedrich Fraundorfer, Horst Bischof
In the second step, we rank the resulting view clusters (i. e. key views with matching partners) according to their impact on the fulfillment of desired quality parameters such as completeness, ground resolution and accuracy.
1 code implementation • 15 Jan 2018 • Michael Opitz, Georg Waltner, Horst Possegger, Horst Bischof
To this end, we divide the last embedding layer of a deep network into an embedding ensemble and formulate training this ensemble as an online gradient boosting problem.
Ranked #13 on Image Retrieval on SOP
no code implementations • ICCV 2017 • Michael Opitz, Georg Waltner, Horst Possegger, Horst Bischof
Learning similarity functions between image pairs with deep neural networks yields highly correlated activations of large embeddings.
no code implementations • CVPR 2017 • Christian Mostegel, Rudolf Prettenthaler, Friedrich Fraundorfer, Horst Bischof
In this paper we present a scalable approach for robustly computing a 3D surface mesh from multi-scale multi-view stereo point clouds that can handle extreme jumps of point density (in our experiments three orders of magnitude).
1 code implementation • 4 Apr 2017 • Gernot Riegler, Ali Osman Ulusoy, Horst Bischof, Andreas Geiger
In this paper, we present a learning based approach to depth fusion, i. e., dense 3D reconstruction from multiple depth images.
no code implementations • 6 Oct 2016 • Georg Poier, Markus Seidl, Matthias Zeppelzauer, Christian Reinbacher, Martin Schaich, Giovanna Bellandi, Alberto Marretta, Horst Bischof
The development of powerful 3D scanning hardware and reconstruction algorithms has strongly promoted the generation of 3D surface reconstructions in different domains.
no code implementations • 1 Sep 2016 • Michael Opitz, Georg Waltner, Georg Poier, Horst Possegger, Horst Bischof
Detection of partially occluded objects is a challenging computer vision problem.
no code implementations • 28 Jul 2016 • Gernot Riegler, David Ferstl, Matthias Rüther, Horst Bischof
In this paper we present a novel method to increase the spatial resolution of depth images.
no code implementations • 27 Jul 2016 • Gernot Riegler, Matthias Rüther, Horst Bischof
We demonstrate that it is feasible to train our method solely on synthetic data that we generate in large quantities for this task.
no code implementations • 6 May 2016 • Christian Mostegel, Markus Rumpler, Friedrich Fraundorfer, Horst Bischof
In this paper we present an autonomous system for acquiring close-range high-resolution images that maximize the quality of a later-on 3D reconstruction with respect to coverage, ground resolution and 3D uncertainty.
no code implementations • CVPR 2016 • Christian Mostegel, Markus Rumpler, Friedrich Fraundorfer, Horst Bischof
Learned confidence measures gain increasing importance for outlier removal and quality improvement in stereo vision.
no code implementations • ICCV 2015 • Gernot Riegler, Samuel Schulter, Matthias Ruther, Horst Bischof
However, this setting is not realistic for practical applications, because the blur is typically different for each test image.
no code implementations • ICCV 2015 • David Ferstl, Matthias Ruther, Horst Bischof
Both the sparse coding and the variational superresolution of the depth are solved based on a primal-dual formulation.
no code implementations • 27 Oct 2015 • Georg Poier, Konstantinos Roditakis, Samuel Schulter, Damien Michel, Horst Bischof, Antonis A. Argyros
Model-based approaches to 3D hand tracking have been shown to perform well in a wide range of scenarios.
no code implementations • CVPR 2015 • Stephan Schraml, Ahmed Nabil Belbachir, Horst Bischof
This paper presents a stereo matching approach for a novel multi-perspective panoramic stereo vision system, making use of asynchronous and non-simultaneous stereo imaging towards real-time 3D 360deg vision.
no code implementations • CVPR 2015 • Horst Possegger, Thomas Mauthner, Horst Bischof
We evaluate our approach on recent online tracking benchmark datasets demonstrating state-of-the-art results.
no code implementations • CVPR 2015 • Thomas Mauthner, Horst Possegger, Georg Waltner, Horst Bischof
We present a novel video saliency detection method to support human activity recognition and weakly supervised training of activity detection algorithms.
no code implementations • CVPR 2015 • Samuel Schulter, Christian Leistner, Horst Bischof
The aim of single image super-resolution is to reconstruct a high-resolution image from a single low-resolution input.
no code implementations • 25 Feb 2015 • Shreyansh Daftry, Christof Hoppe, Horst Bischof
Automatic reconstruction of 3D models from images using multi-view Structure-from-Motion methods has been one of the most fruitful outcomes of computer vision.
no code implementations • CVPR 2014 • Horst Possegger, Thomas Mauthner, Peter M. Roth, Horst Bischof
Robust multi-object tracking-by-detection requires the correct assignment of noisy detection results to object trajectories.
no code implementations • CVPR 2014 • Samuel Schulter, Christian Leistner, Paul Wohlhart, Peter M. Roth, Horst Bischof
In this way, we can simultaneously predict the object probability of a window in a sliding window approach as well as regress its aspect ratio with a single model.
no code implementations • 25 Apr 2014 • Georg Waltner, Thomas Mauthner, Horst Bischof
This paper describes recognition of single player activities in sport with special emphasis on volleyball.
no code implementations • 19 Apr 2014 • Thomas Holzmann, Christof Hoppe, Stefan Kluckner, Horst Bischof
Creating geometric abstracted models from image-based scene reconstructions is difficult due to noise and irregularities in the reconstructed model.
no code implementations • 16 Jan 2014 • Yunjin Chen, Thomas Pock, Horst Bischof
We then introduce an approach to learn both analysis operator and synthesis dictionary simultaneously by using a unified framework of bi-level optimization.
no code implementations • 16 Jan 2014 • Yunjin Chen, Thomas Pock, René Ranftl, Horst Bischof
It is now well known that Markov random fields (MRFs) are particularly effective for modeling image priors in low-level vision.
no code implementations • CVPR 2013 • Samuel Schulter, Paul Wohlhart, Christian Leistner, Amir Saffari, Peter M. Roth, Horst Bischof
Contrary to Boosted Trees, in our method the loss minimization is an inherent part of the tree growing process, thus allowing to keep the benefits of common Random Forests, such as, parallel processing.
no code implementations • CVPR 2013 • Horst Possegger, Sabine Sternig, Thomas Mauthner, Peter M. Roth, Horst Bischof
Combining foreground images from multiple views by projecting them onto a common ground-plane has been recently applied within many multi-object tracking approaches.
no code implementations • CVPR 2013 • Paul Wohlhart, Martin Kostinger, Michael Donoser, Peter M. Roth, Horst Bischof
The development of complex, powerful classifiers and their constant improvement have contributed much to the progress in many fields of computer vision.
no code implementations • CVPR 2013 • Michael Donoser, Horst Bischof
In this paper we revisit diffusion processes on affinity graphs for capturing the intrinsic manifold structure defined by pairwise affinity matrices.
no code implementations • NeurIPS 2012 • Peter Kontschieder, Samuel R. Bulò, Antonio Criminisi, Pushmeet Kohli, Marcello Pelillo, Horst Bischof
In this paper we introduce Context-Sensitive Decision Forests - A new perspective to exploit contextual information in the popular decision forest framework for the object detection problem.