Search Results for author: Konrad Schindler

Found 125 papers, 61 papers with code

Is Continual Learning Ready for Real-world Challenges?

no code implementations15 Feb 2024 Theodora Kontogianni, Yuanwen Yue, Siyu Tang, Konrad Schindler

Our paper aims to initiate a paradigm shift, advocating for the adoption of continual learning methods through new experimental protocols that better emulate real-world conditions to facilitate breakthroughs in the field.

3D Semantic Segmentation Continual Learning

SegmentAnyTree: A sensor and platform agnostic deep learning model for tree segmentation using laser scanning data

no code implementations28 Jan 2024 Maciej Wielgosz, Stefano Puliti, Binbin Xiang, Konrad Schindler, Rasmus Astrup

In conclusion, this study shows the feasibility of a sensor-agnostic model for diverse lidar data, surpassing sensor-specific approaches and setting new standards in tree segmentation, particularly in complex forests.

Instance Segmentation Segmentation +1

Automated forest inventory: analysis of high-density airborne LiDAR point clouds with 3D deep learning

1 code implementation22 Dec 2023 Binbin Xiang, Maciej Wielgosz, Theodora Kontogianni, Torben Peters, Stefano Puliti, Rasmus Astrup, Konrad Schindler

Detailed forest inventories are critical for sustainable and flexible management of forest resources, to conserve various ecosystem services.


Living Scenes: Multi-object Relocalization and Reconstruction in Changing 3D Environments

no code implementations14 Dec 2023 Liyuan Zhu, Shengyu Huang, Konrad Schindler, Iro Armeni

Research into dynamic 3D scene understanding has primarily focused on short-term change tracking from dense observations, while little attention has been paid to long-term changes with sparse observations.

3D Reconstruction Scene Understanding

Dynamic LiDAR Re-simulation using Compositional Neural Fields

no code implementations8 Dec 2023 Hanfeng Wu, Xingxing Zuo, Stefan Leutenegger, Or Litany, Konrad Schindler, Shengyu Huang

We introduce DyNFL, a novel neural field-based approach for high-fidelity re-simulation of LiDAR scans in dynamic driving scenes.

Point2CAD: Reverse Engineering CAD Models from 3D Point Clouds

no code implementations7 Dec 2023 Yujia Liu, Anton Obukhov, Jan Dirk Wegner, Konrad Schindler

Computer-Aided Design (CAD) model reconstruction from point clouds is an important problem at the intersection of computer vision, graphics, and machine learning; it saves the designer significant time when iterating on in-the-wild objects.

CAD Reconstruction Semantic Segmentation

DGInStyle: Domain-Generalizable Semantic Segmentation with Image Diffusion Models and Stylized Semantic Control

no code implementations5 Dec 2023 Yuru Jia, Lukas Hoyer, Shengyu Huang, Tianfu Wang, Luc van Gool, Konrad Schindler, Anton Obukhov

Large, pretrained latent diffusion models (LDMs) have demonstrated an extraordinary ability to generate creative content, specialize to user data through few-shot fine-tuning, and condition their output on other modalities, such as semantic maps.

Autonomous Driving Domain Generalization +1

Neural Fields with Thermal Activations for Arbitrary-Scale Super-Resolution

no code implementations29 Nov 2023 Alexander Becker, Rodrigo Caye Daudt, Nando Metzger, Jan Dirk Wegner, Konrad Schindler

Recent approaches for arbitrary-scale single image super-resolution (ASSR) have used local neural fields to represent continuous signals that can be sampled at different rates.

Image Super-Resolution

High-resolution Population Maps Derived from Sentinel-1 and Sentinel-2

no code implementations23 Nov 2023 Nando Metzger, Rodrigo Caye Daudt, Devis Tuia, Konrad Schindler

With our work we aim to democratize access to up-to-date and high-resolution population maps, recognizing that some regions faced with particularly strong population dynamics may lack the resources for costly micro-census campaigns.

Humanitarian Population Mapping

Nothing Stands Still: A Spatiotemporal Benchmark on 3D Point Cloud Registration Under Large Geometric and Temporal Change

no code implementations15 Nov 2023 Tao Sun, Yan Hao, Shengyu Huang, Silvio Savarese, Konrad Schindler, Marc Pollefeys, Iro Armeni

To this end, we introduce the Nothing Stands Still (NSS) benchmark, which focuses on the spatiotemporal registration of 3D scenes undergoing large spatial and temporal change, ultimately creating one coherent spatiotemporal map.

Point Cloud Registration

Cross-attention Spatio-temporal Context Transformer for Semantic Segmentation of Historical Maps

1 code implementation19 Oct 2023 Sidi Wu, Yizi Chen, Konrad Schindler, Lorenz Hurni

Even though our application is on segmenting historical maps, we believe that the method can be transferred into other fields with similar problems like temporal sequences of satellite images.

Semantic Segmentation Temporal Sequences

Box2Poly: Memory-Efficient Polygon Prediction of Arbitrarily Shaped and Rotated Text

no code implementations20 Sep 2023 Xuyang Chen, Dong Wang, Konrad Schindler, Mingwei Sun, Yongliang Wang, Nicolo Savioli, Liqiu Meng

Recently, Transformer-based text detection techniques have sought to predict polygons by encoding the coordinates of individual boundary vertices using distinct query features.

regression Text Detection

Breathing New Life into 3D Assets with Generative Repainting

2 code implementations15 Sep 2023 Tianfu Wang, Menelaos Kanakis, Konrad Schindler, Luc van Gool, Anton Obukhov

Diffusion-based text-to-image models ignited immense attention from the vision community, artists, and content creators.

Accuracy and Consistency of Space-based Vegetation Height Maps for Forest Dynamics in Alpine Terrain

no code implementations4 Sep 2023 Yuchang Jiang, Marius Rüetschi, Vivien Sainte Fare Garnot, Mauro Marty, Konrad Schindler, Christian Ginzler, Jan D. Wegner

Our results demonstrate that vegetation height maps computed from satellite imagery with deep learning are a valuable, complementary, cost-effective source of evidence to increase the temporal resolution for national forest assessments.

Change Detection

Towards accurate instance segmentation in large-scale LiDAR point clouds

1 code implementation6 Jul 2023 Binbin Xiang, Torben Peters, Theodora Kontogianni, Frawa Vetterli, Stefano Puliti, Rasmus Astrup, Konrad Schindler

Panoptic segmentation is the combination of semantic and instance segmentation: assign the points in a 3D point cloud to semantic categories and partition them into distinct object instances.

Clustering Instance Segmentation +5

SF-FSDA: Source-Free Few-Shot Domain Adaptive Object Detection with Efficient Labeled Data Factory

no code implementations7 Jun 2023 Han Sun, Rui Gong, Konrad Schindler, Luc van Gool

Domain adaptive object detection aims to leverage the knowledge learned from a labeled source domain to improve the performance on an unlabeled target domain.

Object object-detection +2

AGILE3D: Attention Guided Interactive Multi-object 3D Segmentation

no code implementations1 Jun 2023 Yuanwen Yue, Sabarinath Mahadevan, Jonas Schult, Francis Engelmann, Bastian Leibe, Konrad Schindler, Theodora Kontogianni

In an iterative process, the model assigns each data point to an object (or the background), while the user corrects errors in the resulting segmentation and feeds them back into the model.

Binary Classification Interactive Segmentation +2

Mixture of Experts with Uncertainty Voting for Imbalanced Deep Regression Problems

no code implementations24 May 2023 Yuchang Jiang, Vivien Sainte Fare Garnot, Konrad Schindler, Jan Dirk Wegner

For regression, recent work relies on the continuity of the distribution; whereas for classification there has been a trend to employ mixture-of-expert models and let some ensemble members specialize in predictions for the sparser regions.

Probabilistic Deep Learning regression

U-TILISE: A Sequence-to-sequence Model for Cloud Removal in Optical Satellite Time Series

1 code implementation22 May 2023 Corinne Stucker, Vivien Sainte Fare Garnot, Konrad Schindler

Satellite image time series in the optical and infrared spectrum suffer from frequent data gaps due to cloud cover, cloud shadows, and temporary sensor outages.

Cloud Removal Representation Learning +1

Neural LiDAR Fields for Novel View Synthesis

no code implementations ICCV 2023 Shengyu Huang, Zan Gojcic, Zian Wang, Francis Williams, Yoni Kasten, Sanja Fidler, Konrad Schindler, Or Litany

We present Neural Fields for LiDAR (NFL), a method to optimise a neural field scene representation from LiDAR measurements, with the goal of synthesizing realistic LiDAR scans from novel viewpoints.

Novel View Synthesis Semantic Segmentation

A Review of Panoptic Segmentation for Mobile Mapping Point Clouds

1 code implementation27 Apr 2023 Binbin Xiang, Yuanwen Yue, Torben Peters, Konrad Schindler

Moreover, a modular pipeline is set up to perform comprehensive, systematic experiments to assess the state of panoptic segmentation in the context of street mapping.

Instance Segmentation Panoptic Segmentation +2

DEFLOW: Self-supervised 3D Motion Estimation of Debris Flow

no code implementations5 Apr 2023 Liyuan Zhu, Yuru Jia, Shengyu Huang, Nicholas Meyer, Andreas Wieser, Konrad Schindler, Jordan Aaron

Our model achieves state-of-the-art optical flow and depth estimation on our dataset, and fully automates the motion estimation for debris flows.

Autonomous Driving Depth Estimation +4

BiasBed - Rigorous Texture Bias Evaluation

1 code implementation CVPR 2023 Nikolai Kalischek, Rodrigo Caye Daudt, Torben Peters, Reinhard Furrer, Jan D. Wegner, Konrad Schindler

With the release of BiasBed, we hope to foster a common understanding of consistent and meaningful comparisons, and consequently faster progress towards learning methods free of texture bias.

Model Selection

Context-aware multi-head self-attentional neural network model for next location prediction

2 code implementations4 Dec 2022 Ye Hong, Yatao Zhang, Konrad Schindler, Martin Raubal

Accurate activity location prediction is a crucial component of many mobility applications and is particularly required to develop personalized, sustainable transportation systems.

Connecting the Dots: Floorplan Reconstruction Using Two-Level Queries

1 code implementation CVPR 2023 Yuanwen Yue, Theodora Kontogianni, Konrad Schindler, Francis Engelmann

Instead, we formulate floorplan reconstruction as a single-stage structured prediction task: find a variable-size set of polygons, which in turn are variable-length sequences of ordered vertices.

Structured Prediction Vocal Bursts Valence Prediction

BiasBed -- Rigorous Texture Bias Evaluation

1 code implementation23 Nov 2022 Nikolai Kalischek, Rodrigo C. Daudt, Torben Peters, Reinhard Furrer, Jan D. Wegner, Konrad Schindler

With the release of BiasBed, we hope to foster a common understanding of consistent and meaningful comparisons, and consequently faster progress towards learning methods free of texture bias.

Model Selection

Guided Depth Super-Resolution by Deep Anisotropic Diffusion

1 code implementation CVPR 2023 Nando Metzger, Rodrigo Caye Daudt, Konrad Schindler

In this work, we propose a novel approach which combines guided anisotropic diffusion with a deep convolutional network and advances the state of the art for guided depth super-resolution.


TT-NF: Tensor Train Neural Fields

1 code implementation30 Sep 2022 Anton Obukhov, Mikhail Usvyatsov, Christos Sakaridis, Konrad Schindler, Luc van Gool

Learning neural fields has been an active topic in deep learning research, focusing, among other issues, on finding more compact and easy-to-fit representations.

Denoising Low-rank compression

T4DT: Tensorizing Time for Learning Temporal 3D Visual Data

2 code implementations2 Aug 2022 Mikhail Usvyatsov, Rafael Ballester-Rippoll, Lina Bashaeva, Konrad Schindler, Gonzalo Ferrer, Ivan Oseledets

We show that low-rank tensor compression is extremely compact to store and query time-varying signed distance functions.

Dynamic 3D Scene Analysis by Point Cloud Accumulation

1 code implementation25 Jul 2022 Shengyu Huang, Zan Gojcic, Jiahui Huang, Andreas Wieser, Konrad Schindler

Compared to state-of-the-art scene flow estimators, our proposed approach aims to align all 3D points in a common reference frame correctly accumulating the points on the individual objects.

Autonomous Vehicles Semantic Segmentation +1

tntorch: Tensor Network Learning with PyTorch

1 code implementation22 Jun 2022 Mikhail Usvyatsov, Rafael Ballester-Ripoll, Konrad Schindler

We present tntorch, a tensor learning framework that supports multiple decompositions (including Candecomp/Parafac, Tucker, and Tensor Train) under a unified interface.

Recognition of Unseen Bird Species by Learning from Field Guides

1 code implementation3 Jun 2022 Andrés C. Rodríguez, Stefano D'Aronco, Rodrigo Caye Daudt, Jan D. Wegner, Konrad Schindler

Illustrations contained in field guides deliberately focus on discriminative properties of each species, and can serve as side information to transfer knowledge from seen to unseen bird species.

Generalized Zero-Shot Learning

FiLM-Ensemble: Probabilistic Deep Learning via Feature-wise Linear Modulation

1 code implementation31 May 2022 Mehmet Ozgur Turkoglu, Alexander Becker, Hüseyin Anil Gündüz, Mina Rezaei, Bernd Bischl, Rodrigo Caye Daudt, Stefano D'Aronco, Jan Dirk Wegner, Konrad Schindler

We show that the idea can be extended to uncertainty quantification: by modulating the network activations of a single deep network with FiLM, one obtains a model ensemble with high diversity, and consequently well-calibrated estimates of epistemic uncertainty, with low computational overhead in comparison.

Multi-Task Learning Probabilistic Deep Learning +1

Urban Change Forecasting from Satellite Images

no code implementations27 Apr 2022 Nando Metzger, Mehmet Özgür Türkoglu, Rodrigo Caye Daudt, Jan Dirk Wegner, Konrad Schindler

In Stage 1, a U-Net backbone is pretrained within a Siamese network architecture that aims to solve a (building) change detection task.

Change Detection Management

Interactive Object Segmentation in 3D Point Clouds

no code implementations14 Apr 2022 Theodora Kontogianni, Ekin Celikkan, Siyu Tang, Konrad Schindler

We propose an interactive approach for 3D instance segmentation, where users can iteratively collaborate with a deep learning model to segment objects in a 3D point cloud directly.

3D Instance Segmentation Image Segmentation +4

A high-resolution canopy height model of the Earth

no code implementations13 Apr 2022 Nico Lang, Walter Jetz, Konrad Schindler, Jan Dirk Wegner

The worldwide variation in vegetation height is fundamental to the global carbon cycle and central to the functioning of ecosystems and their biodiversity.

Decision Making Probabilistic Deep Learning +2

BARC: Learning to Regress 3D Dog Shape from Images by Exploiting Breed Information

no code implementations CVPR 2022 Nadine Rueegg, Silvia Zuffi, Konrad Schindler, Michael J. Black

But, even with a better shape model, the problem of regressing dog shape from an image is still challenging because we lack paired images with 3D ground truth.

Learning Graph Regularisation for Guided Super-Resolution

1 code implementation CVPR 2022 Riccardo de Lutio, Alexander Becker, Stefano D'Aronco, Stefania Russo, Jan D. Wegner, Konrad Schindler

With the decision to employ the source as a constraint rather than only as an input to the prediction, our method differs from state-of-the-art deep architectures for guided super-resolution, which produce targets that, when downsampled, will only approximately reproduce the source.


ImpliCity: City Modeling from Satellite Images with Deep Implicit Occupancy Fields

1 code implementation24 Jan 2022 Corinne Stucker, Bingxin Ke, Yuanwen Yue, Shengyu Huang, Iro Armeni, Konrad Schindler

To make full use of the point cloud and the underlying images, we introduce ImpliCity, a neural representation of the 3D scene as an implicit, continuous occupancy field, driven by learned embeddings of the point cloud and a stereo pair of ortho-photos.

3D Reconstruction from public webcams

no code implementations21 Aug 2021 Tianyu Wu, Konrad Schindler, Cenek Albl

It turns out that the task to reconstruct scene structure from webcam streams is very different from standard structure-from-motion (SfM), and conventional SfM pipelines fail.

3D Reconstruction 3D Scene Reconstruction

Mapping Vulnerable Populations with AI

no code implementations29 Jul 2021 Benjamin Kellenberger, John E. Vargas-Muñoz, Devis Tuia, Rodrigo C. Daudt, Konrad Schindler, Thao T-T Whelan, Brenda Ayo, Ferda Ofli, Muhammad Imran

Building functions shall be retrieved by parsing social media data like for instance tweets, as well as ground-based imagery, to automatically identify different buildings functions and retrieve further information such as the number of building stories.

Humanitarian Image Segmentation +1

Digital Taxonomist: Identifying Plant Species in Community Scientists' Photographs

no code implementations7 Jun 2021 Riccardo de Lutio, Yihang She, Stefano D'Aronco, Stefania Russo, Philipp Brun, Jan D. Wegner, Konrad Schindler

Automatic identification of plant specimens from amateur photographs could improve species range maps, thus supporting ecosystems research as well as conservation efforts.

Multimodal Deep Learning

Mapping oil palm density at country scale: An active learning approach

no code implementations24 May 2021 Andrés C. Rodríguez, Stefano D'Aronco, Konrad Schindler, Jan D. Wegner

To that end, we propose a new, active deep learning method to estimate oil palm density at large scale from Sentinel-2 satellite images, and apply it to generate complete maps for Malaysia and Indonesia.

Active Learning Density Estimation

Visual Camera Re-Localization Using Graph Neural Networks and Relative Pose Supervision

1 code implementation6 Apr 2021 Mehmet Ozgur Turkoglu, Eric Brachmann, Konrad Schindler, Gabriel Brostow, Aron Monszpart

Visual re-localization means using a single image as input to estimate the camera's location and orientation relative to a pre-recorded environment.


Global canopy height regression and uncertainty estimation from GEDI LIDAR waveforms with deep ensembles

1 code implementation5 Mar 2021 Nico Lang, Nikolai Kalischek, John Armston, Konrad Schindler, Ralph Dubayah, Jan Dirk Wegner

NASA's Global Ecosystem Dynamics Investigation (GEDI) is a key climate mission whose goal is to advance our understanding of the role of forests in the global carbon cycle.

Probabilistic Deep Learning regression

PC2WF: 3D Wireframe Reconstruction from Raw Point Clouds

1 code implementation ICLR 2021 Yujia Liu, Stefano D'Aronco, Konrad Schindler, Jan Dirk Wegner

Next, the corners are linked with an exhaustive set of candidate edges, which is again pruned to obtain the final wireframe.

3D Wireframe Reconstruction

Walk2Map: Extracting Floor Plans from Indoor Walk Trajectories

no code implementations27 Feb 2021 Claudio Mura, Renato Pajarola, Konrad Schindler, Niloy Mitra

Thanks to recent advances in data-driven inertial odometry, such minimalistic input data can be acquired from the IMU readings of consumer-level smartphones, which allows for an effortless and scalable mapping of real-world indoor spaces.

Image-to-Image Translation Management

Crop mapping from image time series: deep learning with multi-scale label hierarchies

1 code implementation17 Feb 2021 Mehmet Ozgur Turkoglu, Stefano D'Aronco, Gregor Perich, Frank Liebisch, Constantin Streit, Konrad Schindler, Jan Dirk Wegner

The three-level label hierarchy is encoded in a convolutional, recurrent neural network (convRNN), such that for each pixel the model predicts three labels at different level of granularity.

Crop Classification General Classification +2

Crop Classification under Varying Cloud Cover with Neural Ordinary Differential Equations

1 code implementation4 Dec 2020 Nando Metzger, Mehmet Ozgur Turkoglu, Stefano D'Aronco, Jan Dirk Wegner, Konrad Schindler

We propose to use neural ordinary differential equations (NODEs) in combination with RNNs to classify crop types in irregularly spaced image sequences.

Crop Classification General Classification +1

MOTChallenge: A Benchmark for Single-Camera Multiple Target Tracking

no code implementations15 Oct 2020 Patrick Dendorfer, Aljoša Ošep, Anton Milan, Konrad Schindler, Daniel Cremers, Ian Reid, Stefan Roth, Laura Leal-Taixé

We present MOTChallenge, a benchmark for single-camera Multiple Object Tracking (MOT) launched in late 2014, to collect existing and new data, and create a framework for the standardized evaluation of multiple object tracking methods.

Multiple Object Tracking Multiple People Tracking +3

Deep Active Learning in Remote Sensing for data efficient Change Detection

1 code implementation25 Aug 2020 Vít Růžička, Stefano D'Aronco, Jan Dirk Wegner, Konrad Schindler

We investigate active learning in the context of deep neural network models for change detection and map updating.

Active Learning Change Detection

KAPLAN: A 3D Point Descriptor for Shape Completion

no code implementations31 Jul 2020 Audrey Richard, Ian Cherabier, Martin R. Oswald, Marc Pollefeys, Konrad Schindler

We present a novel 3D shape completion method that operates directly on unstructured point clouds, thus avoiding resource-intensive data structures like voxel grids.

3D Shape Reconstruction

From two rolling shutters to one global shutter

no code implementations CVPR 2020 Cenek Albl, Zuzana Kukelova, Viktor Larsson, Tomas Pajdla, Konrad Schindler

Most consumer cameras are equipped with electronic rolling shutter, leading to image distortions when the camera moves during image capture.

Vocal Bursts Valence Prediction

Photometric Multi-View Mesh Refinement for High-Resolution Satellite Images

no code implementations10 May 2020 Mathias Rothermel, Ke Gong, Dieter Fritsch, Konrad Schindler, Norbert Haala

Modern high-resolution satellite sensors collect optical imagery with ground sampling distances (GSDs) of 30-50cm, which has sparked a renewed interest in photogrammetric 3D surface reconstruction from satellite data.

Surface Reconstruction Vocal Bursts Intensity Prediction

Fine-grained Species Recognition with Privileged Pooling: Better Sample Efficiency Through Supervised Attention

1 code implementation20 Mar 2020 Andres C. Rodriguez, Stefano D'Aronco, Konrad Schindler, Jan Dirk Wegner

We propose a scheme for supervised image classification that uses privileged information, in the form of keypoint annotations for the training data, to learn strong models from small and/or biased training sets.

Image Classification

MOT20: A benchmark for multi object tracking in crowded scenes

1 code implementation19 Mar 2020 Patrick Dendorfer, Hamid Rezatofighi, Anton Milan, Javen Shi, Daniel Cremers, Ian Reid, Stefan Roth, Konrad Schindler, Laura Leal-Taixé

The benchmark for Multiple Object Tracking, MOTChallenge, was launched with the goal to establish a standardized evaluation of multiple object tracking methods.

Multi-Object Tracking Multiple Object Tracking with Transformer +2

Reconstruction of 3D flight trajectories from ad-hoc camera networks

2 code implementations10 Mar 2020 Jingtong Li, Jesse Murray, Dorina Ismaili, Konrad Schindler, Cenek Albl

We present a method to reconstruct the 3D trajectory of an airborne robotic system only from videos recorded with cameras that are unsynchronized, may feature rolling shutter distortion, and whose viewpoints are unknown.

Indoor Scene Recognition in 3D

2 code implementations28 Feb 2020 Shengyu Huang, Mikhail Usvyatsov, Konrad Schindler

Moreover, we advocate multi-task learning as a way of improving scene recognition, building on the fact that the scene type is highly correlated with the objects in the scene, and therefore with its semantic segmentation into different object classes.

Multi-Task Learning Scene Recognition +1

Photi-LakeIce Dataset

1 code implementation ISPRS Congress 2020 Rajanie Prabha, Manu Tom, Mathias Rothermel, Emmanuel Baltsavias, Laura Leal-Taixe, Konrad Schindler

On average, it achieves intersection-over-union (IoU) values of ~71% across different cameras and ~69% across different winters, greatly outperforming prior work.

Change detection for remote sensing images Image Segmentation +4

Lake Ice Monitoring with Webcams and Crowd-Sourced Images

2 code implementations18 Feb 2020 Rajanie Prabha, Manu Tom, Mathias Rothermel, Emmanuel Baltsavias, Laura Leal-Taixe, Konrad Schindler

On average, it achieves intersection-over-union (IoU) values of ~71% across different cameras and ~69% across different winters, greatly outperforming prior work.

Change detection for remote sensing images Image Segmentation +4

Lake Ice Detection from Sentinel-1 SAR with Deep Learning

1 code implementation17 Feb 2020 Manu Tom, Roberto Aguilar, Pascal Imhof, Silvan Leinss, Emmanuel Baltsavias, Konrad Schindler

Lake ice, as part of the Essential Climate Variable (ECV) lakes, is an important indicator to monitor climate change and global warming.

Change detection for remote sensing images Lake Ice Monitoring +2

ResDepth: Learned Residual Stereo Reconstruction

1 code implementation22 Jan 2020 Corinne Stucker, Konrad Schindler

We propose an embarrassingly simple but very effective scheme for high-quality dense stereo reconstruction: (i) generate an approximate reconstruction with your favourite stereo matcher; (ii) rewarp the input images with that approximate model; (iii) with the initial reconstruction and the warped images as input, train a deep network to enhance the reconstruction by regressing a residual correction; and (iv) if desired, iterate the refinement with the new, improved reconstruction.

Learned Multi-View Texture Super-Resolution

no code implementations14 Jan 2020 Audrey Richard, Ian Cherabier, Martin R. Oswald, Vagia Tsiminaki, Marc Pollefeys, Konrad Schindler

We present a super-resolution method capable of creating a high-resolution texture map for a virtual 3D object from a set of lower-resolution images of that object.

Image Super-Resolution

Self-supervised Object Motion and Depth Estimation from Video

no code implementations9 Dec 2019 Qi Dai, Vaishakh Patil, Simon Hecker, Dengxin Dai, Luc van Gool, Konrad Schindler

We present a self-supervised learning framework to estimate the individual object motion and monocular depth from video.

Depth Estimation Instance Segmentation +5

Gating Revisited: Deep Multi-layer RNNs That Can Be Trained

3 code implementations25 Nov 2019 Mehmet Ozgur Turkoglu, Stefano D'Aronco, Jan Dirk Wegner, Konrad Schindler

We propose a new STAckable Recurrent cell (STAR) for recurrent neural networks (RNNs), which has fewer parameters than widely used LSTM and GRU while being more robust against vanishing or exploding gradients.

Action Recognition In Videos Language Modelling +2

From Google Maps to a Fine-Grained Catalog of Street trees

no code implementations7 Oct 2019 Steve Branson, Jan Dirk Wegner, David Hall, Nico Lang, Konrad Schindler, Pietro Perona

We believe this is the first work to exploit publicly available image data for fine-grained tree mapping at city-scale, respectively over many thousands of trees.

Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer

14 code implementations2 Jul 2019 René Ranftl, Katrin Lasinger, David Hafner, Konrad Schindler, Vladlen Koltun

In particular, we propose a robust training objective that is invariant to changes in depth range and scale, advocate the use of principled multi-objective learning to combine data from different sources, and highlight the importance of pretraining encoders on auxiliary tasks.

Monocular Depth Estimation

Country-wide high-resolution vegetation height mapping with Sentinel-2

no code implementations30 Apr 2019 Nico Lang, Konrad Schindler, Jan Dirk Wegner

Sentinel-2 multi-spectral images collected over periods of several months were used to estimate vegetation height for Gabon and Switzerland.

Surface Reconstruction Vocal Bursts Intensity Prediction

Guided Super-Resolution as Pixel-to-Pixel Transformation

2 code implementations ICCV 2019 Riccardo de Lutio, Stefano D'Aronco, Jan Dirk Wegner, Konrad Schindler

Guided super-resolution is a unifying framework for several computer vision tasks where the inputs are a low-resolution source image of some target quantity (e. g., perspective depth acquired with a time-of-flight camera) and a high-resolution guide image from a different domain (e. g., a grey-scale image from a conventional camera); and the target output is a high-resolution version of the source (in our example, a high-res depth map).


Visual recognition in the wild by sampling deep similarity functions

no code implementations15 Mar 2019 Mikhail Usvyatsov, Konrad Schindler

However, a robot moving in the wild, i. e., in an environment that is not known at the time the recognition system is trained, will often face \emph{domain shift}: the training data cannot be assumed to exhaustively cover all the within-class variability that will be encountered in the test data.

Object Recognition Test

Variational 3D-PIV with Sparse Descriptors

no code implementations9 Apr 2018 Katrin Lasinger, Christoph Vogel, Thomas Pock, Konrad Schindler

We propose a new method for iterative particle reconstruction (IPR), in which the locations and intensities of all particles are inferred in one joint energy minimization.

3D Fluid Flow Estimation with Integrated Particle Reconstruction

1 code implementation9 Apr 2018 Katrin Lasinger, Christoph Vogel, Thomas Pock, Konrad Schindler

We show, for the first time, how to jointly reconstruct both the individual tracer particles and a dense 3D fluid motion field from the image data, using an integrated energy minimization.

3D Reconstruction Motion Estimation

Super-resolution of Sentinel-2 images: Learning a globally applicable deep neural network

3 code implementations12 Mar 2018 Charis Lanaras, José Bioucas-Dias, Silvano Galliani, Emmanuel Baltsavias, Konrad Schindler

The aim of this research is to super-resolve the lower-resolution (20 m and 60 m Ground Sampling Distance - GSD) bands to 10 m GSD, so as to obtain a complete data cube at the maximal sensor resolution.


Inference, Learning and Attention Mechanisms that Exploit and Preserve Sparsity in Convolutional Networks

1 code implementation31 Jan 2018 Timo Hackel, Mikhail Usvyatsov, Silvano Galliani, Jan D. Wegner, Konrad Schindler

While CNNs naturally lend themselves to densely sampled data, and sophisticated implementations are available, they lack the ability to efficiently process sparse data.

Semantic 3D Reconstruction with Finite Element Bases

no code implementations4 Oct 2017 Audrey Richard, Christoph Vogel, Maros Blaha, Thomas Pock, Konrad Schindler

We propose a novel framework for the discretisation of multi-label problems on arbitrary, continuous domains.

3D Reconstruction valid

Learning Aerial Image Segmentation from Online Maps

2 code implementations21 Jul 2017 Pascal Kaiser, Jan Dirk Wegner, Aurelien Lucchi, Martin Jaggi, Thomas Hofmann, Konrad Schindler

We adapt a state-of-the-art CNN architecture for semantic segmentation of buildings and roads in aerial images, and compare its performance when using different training data sets, ranging from manually labeled, pixel-accurate ground truth of the same city to automatic training data derived from OpenStreetMap data from distant locations.

General Classification Image Segmentation +2 A new Large-scale Point Cloud Classification Benchmark

1 code implementation12 Apr 2017 Timo Hackel, Nikolay Savinov, Lubor Ladicky, Jan D. Wegner, Konrad Schindler, Marc Pollefeys

With the massive data set presented in this paper, we aim at closing this data gap to help unleash the full potential of deep learning methods for 3D labelling tasks.

3D Point Cloud Classification Classification +5

Learned Multi-Patch Similarity

1 code implementation ICCV 2017 Wilfried Hartmann, Silvano Galliani, Michal Havlena, Luc van Gool, Konrad Schindler

Estimating a depth map from multiple views of a scene is a fundamental task in computer vision.

Classification With an Edge: Improving Semantic Image Segmentation with Boundary Detection

1 code implementation5 Dec 2016 Dimitrios Marmanis, Konrad Schindler, Jan Dirk Wegner, Silvano Galliani, Mihai Datcu, Uwe Stilla

We present an end-to-end trainable deep convolutional neural network (DCNN) for semantic segmentation with built-in awareness of semantically meaningful boundaries.

Boundary Detection Edge Detection +4

Tracking with multi-level features

no code implementations25 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.

Clustering Multiple Object Tracking

Large-Scale Location Recognition and the Geometric Burstiness Problem

1 code implementation CVPR 2016 Torsten Sattler, Michal Havlena, Konrad Schindler, Marc Pollefeys

Visual location recognition is the task of determining the place depicted in a query image from a given database of geo-tagged images.

Image Retrieval Re-Ranking +1

Cataloging Public Objects Using Aerial and Street-Level Images - Urban Trees

no code implementations CVPR 2016 Jan D. Wegner, Steven Branson, David Hall, Konrad Schindler, Pietro Perona

The main technical challenge is combining test time information from multiple views of each geographic location (e. g., aerial and street views).


Just Look at the Image: Viewpoint-Specific Surface Normal Prediction for Improved Multi-View Reconstruction

no code implementations CVPR 2016 Silvano Galliani, Konrad Schindler

By training from known points in the same image, the prediction is specifically tailored to the materials and lighting conditions of the particular scene, as well as to the precise camera viewpoint.

Learning by tracking: Siamese CNN for robust target association

no code implementations26 Apr 2016 Laura Leal-Taixé, Cristian Canton Ferrer, Konrad Schindler

This paper introduces a novel approach to the task of data association within the context of pedestrian tracking, by introducing a two-stage learning scheme to match pairs of detections.

Multiple People Tracking Optical Flow Estimation +1

Online Multi-Target Tracking Using Recurrent Neural Networks

no code implementations13 Apr 2016 Anton Milan, Seyed Hamid Rezatofighi, Anthony Dick, Ian Reid, Konrad Schindler

Here, we propose for the first time, an end-to-end learning approach for online multi-target tracking.

MOT16: A Benchmark for Multi-Object Tracking

8 code implementations2 Mar 2016 Anton Milan, Laura Leal-Taixe, Ian Reid, Stefan Roth, Konrad Schindler

Recently, a new benchmark for Multiple Object Tracking, MOTChallenge, was launched with the goal of collecting existing and new data and creating a framework for the standardized evaluation of multiple object tracking methods.

Multi-Object Tracking Multiple Object Tracking +2

Hyperspectral Super-Resolution by Coupled Spectral Unmixing

no code implementations ICCV 2015 Charis Lanaras, Emmanuel Baltsavias, Konrad Schindler

Hyperspectral super-resolution addresses this problem, by fusing a low-resolution hyperspectral image and a conventional high-resolution image into a product of both high spatial and high spectral resolution.


Joint Tracking and Segmentation of Multiple Targets

no code implementations CVPR 2015 Anton Milan, Laura Leal-Taixe, Konrad Schindler, Ian Reid

Tracking-by-detection has proven to be the most successful strategy to address the task of tracking multiple targets in unconstrained scenarios.

Video Segmentation Video Semantic Segmentation

MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking

2 code implementations8 Apr 2015 Laura Leal-Taixé, Anton Milan, Ian Reid, Stefan Roth, Konrad Schindler

We discuss the challenges of creating such a framework, collecting existing and new data, gathering state-of-the-art methods to be tested on the datasets, and finally creating a unified evaluation system.

3D Reconstruction Multiple Object Tracking +3

Towards Scene Understanding with Detailed 3D Object Representations

no code implementations18 Nov 2014 M. Zeeshan Zia, Michael Stark, Konrad Schindler

An object class - in our case cars - is modeled as a deformable 3D wireframe, which enables fine-grained modeling at the level of individual vertices and faces.

3D Pose Estimation Object +3

Predicting Matchability

no code implementations CVPR 2014 Wilfried Hartmann, Michal Havlena, Konrad Schindler

The initial steps of many computer vision algorithms are interest point extraction and matching.

Explicit Occlusion Modeling for 3D Object Class Representations

no code implementations CVPR 2013 M. Zeeshan Zia, Michael Stark, Konrad Schindler

In this paper, we tackle the challenge of modeling occlusion in the context of a 3D geometric object class model that is capable of fine-grained, part-level 3D object reconstruction.

3D Object Reconstruction Object +2

Detection- and Trajectory-Level Exclusion in Multiple Object Tracking

no code implementations CVPR 2013 Anton Milan, Konrad Schindler, Stefan Roth

When tracking multiple targets in crowded scenarios, modeling mutual exclusion between distinct targets becomes important at two levels: (1) in data association, each target observation should support at most one trajectory and each trajectory should be assigned at most one observation per frame; (2) in trajectory estimation, two trajectories should remain spatially separated at all times to avoid collisions.

Multiple Object Tracking Object

Cannot find the paper you are looking for? You can Submit a new open access paper.