no code implementations • 13 Apr 2023 • Denys Rozumnyi, Jiri Matas, Marc Pollefeys, Vittorio Ferrari, Martin R. Oswald
We argue that this representation is limited and instead propose to guide and improve 2D tracking with an explicit object representation, namely the textured 3D shape and 6DoF pose in each video frame.
1 code implementation • 23 Mar 2023 • Tomas Vojir, Jan Sochman, Rahaf Aljundi, Jiri Matas
In this work, we take a different approach and propose to leverage generic pre-trained representations.
no code implementations • 17 Mar 2023 • David Korcak, Jiri Matas
The approach exploits the relation of the exposure fraction, optical flow, and linear motion blur.
no code implementations • 25 Feb 2023 • Martin Sundermeyer, Tomas Hodan, Yann Labbe, Gu Wang, Eric Brachmann, Bertram Drost, Carsten Rother, Jiri Matas
In 2022, we witnessed another significant improvement in the pose estimation accuracy -- the state of the art, which was 56. 9 AR$_C$ in 2019 (Vidal et al.) and 69. 8 AR$_C$ in 2020 (CosyPose), moved to new heights of 83. 7 AR$_C$ (GDRNPP).
2 code implementations • 20 Feb 2023 • Daniel Barath, Dmytro Mishkin, Michal Polic, Wolfgang Förstner, Jiri Matas
We present a large-scale dataset of Planes in 3D, Pi3D, of roughly 1000 planes observed in 10 000 images from the 1DSfM dataset, and HEB, a large-scale homography estimation benchmark leveraging Pi3D.
1 code implementation • 24 Jan 2023 • Jonas Serych, Jiri Matas
We propose WOFT -- a novel method for planar object tracking that estimates a full 8 degrees-of-freedom pose, i. e. the homography w. r. t.
no code implementations • CVPR 2023 • Daniel Barath, Dmytro Mishkin, Michal Polic, Wolfgang Förstner, Jiri Matas
We present a large-scale dataset of Planes in 3D, Pi3D, of roughly 1000 planes observed in 10 000 images from the 1DSfM dataset, and HEB, a large-scale homography estimation benchmark leveraging Pi3D.
1 code implementation • 26 Dec 2022 • Tong Wei, Yash Patel, Alexander Shekhovtsov, Jiri Matas, Daniel Barath
We propose $\nabla$-RANSAC, a generalized differentiable RANSAC that allows learning the entire randomized robust estimation pipeline.
1 code implementation • 7 Oct 2022 • Alan Lukezic, Ziga Trojer, Jiri Matas, Matej Kristan
Visual object tracking has focused predominantly on opaque objects, while transparent object tracking received very little attention.
no code implementations • 9 Aug 2022 • Phong Nguyen-Ha, Lam Huynh, Esa Rahtu, Jiri Matas, Janne Heikkila
We present HRF-Net, a novel view synthesis method based on holistic radiance fields that renders novel views using a set of sparse inputs.
no code implementations • 15 Jul 2022 • Jan Docekal, Jakub Rozlivek, Jiri Matas, Matej Hoffmann
In particular, (i) we survey existing datasets with human pose annotation from the perspective of close proximity images and prepare and make publicly available a new Human in Close Proximity (HiCP) dataset; (ii) we quantitatively and qualitatively compare state-of-the-art human whole-body 2D keypoint detection methods (OpenPose, MMPose, AlphaPose, Detectron2) on this dataset; (iii) since accurate detection of hands and fingers is critical in applications with handovers, we evaluate the performance of the MediaPipe hand detector; (iv) we deploy the algorithms on a humanoid robot with an RGB-D camera on its head and evaluate the performance in 3D human keypoint detection.
no code implementations • 3 Mar 2022 • Lam Huynh, Esa Rahtu, Jiri Matas, Janne Heikkila
We present LDP, a lightweight dense prediction neural architecture search (NAS) framework.
no code implementations • 6 Dec 2021 • Sajid Javed, Martin Danelljan, Fahad Shahbaz Khan, Muhammad Haris Khan, Michael Felsberg, Jiri Matas
Accurate and robust visual object tracking is one of the most challenging and fundamental computer vision problems.
no code implementations • 28 Nov 2021 • Wei Tong, Jiri Matas, Daniel Barath
We propose Deep MAGSAC++ combining the advantages of traditional and deep robust estimators.
no code implementations • CVPR 2022 • Xiaoyan Xing, Yanlin Qian, Sibo Feng, Yuhan Dong, Jiri Matas
In this paper, we present Point Cloud Color Constancy, in short PCCC, an illumination chromaticity estimation algorithm exploiting a point cloud.
no code implementations • 6 Sep 2021 • Dengxin Dai, Arun Balajee Vasudevan, Jiri Matas, Luc van Gool
Humans can robustly recognize and localize objects by using visual and/or auditory cues.
2 code implementations • CVPR 2022 • Yash Patel, Giorgos Tolias, Jiri Matas
This work focuses on learning deep visual representation models for retrieval by exploring the interplay between a new loss function, the batch size, and a new regularization approach.
Ranked #1 on Vehicle Re-Identification on VehicleID Small
no code implementations • 25 Aug 2021 • Lam Huynh, Matteo Pedone, Phong Nguyen, Jiri Matas, Esa Rahtu, Janne Heikkila
In addition, we introduce a normalized Hessian loss term invariant to scaling and shear along the depth direction, which is shown to substantially improve the accuracy.
no code implementations • 25 Aug 2021 • Lam Huynh, Phong Nguyen, Jiri Matas, Esa Rahtu, Janne Heikkila
This paper presents a novel neural architecture search method, called LiDNAS, for generating lightweight monocular depth estimation models.
1 code implementation • 22 Jun 2021 • Tomas Sipka, Milan Sulc, Jiri Matas
In many computer vision classification tasks, class priors at test time often differ from priors on the training set.
no code implementations • ICCV 2021 • Maksym Ivashechkin, Daniel Barath, Jiri Matas
Experiments on four standard datasets show that VSAC is significantly faster than all its predecessors and runs on average in 1-2 ms, on a CPU.
no code implementations • 11 Apr 2021 • Maksym Ivashechkin, Daniel Barath, Jiri Matas
We review the most recent RANSAC-like hypothesize-and-verify robust estimators.
1 code implementation • CVPR 2023 • Daniel Barath, Denys Rozumny, Ivan Eichhardt, Levente Hajder, Jiri Matas
Dominant instances are found via a RANSAC-like sampling and a consolidation process driven by a model quality function considering previously proposed instances.
no code implementations • 8 Mar 2021 • Yash Patel, Jiri Matas
This paper proposes a procedure to train a scene text recognition model using a robust learned surrogate of edit distance.
1 code implementation • ICCV 2021 • Tomas Vojir, Tomas Sipka, Rahaf Aljundi, Nikolay Chumerin, Daniel Olmeda Reino, Jiri Matas
To that end, we propose a reconstruction module that can be used with many existing semantic segmentation networks, and that is trained to recognize and reconstruct road (drivable) surface from a small bottleneck.
no code implementations • ICCV 2021 • Lam Huynh, Phong Nguyen, Jiri Matas, Esa Rahtu, Janne Heikkila
In this paper, we propose enhancing monocular depth estimation by adding 3D points as depth guidance.
1 code implementation • ICCV 2021 • Denys Rozumnyi, Jiri Matas, Filip Sroubek, Marc Pollefeys, Martin R. Oswald
Compared to other methods, such as deblatting, the inference is of several orders of magnitude faster and allows applications such as real-time fast moving object detection and retrieval in large video collections.
5 code implementations • CVPR 2021 • Denys Rozumnyi, Martin R. Oswald, Vittorio Ferrari, Jiri Matas, Marc Pollefeys
We propose a method that, given a single image with its estimated background, outputs the object's appearance and position in a series of sub-frames as if captured by a high-speed camera (i. e. temporal super-resolution).
Ranked #1 on Video Super-Resolution on Falling Objects
no code implementations • 29 Nov 2020 • Phong Nguyen, Animesh Karnewar, Lam Huynh, Esa Rahtu, Jiri Matas, Janne Heikkila
We propose a new cascaded architecture for novel view synthesis, called RGBD-Net, which consists of two core components: a hierarchical depth regression network and a depth-aware generator network.
1 code implementation • CVPR 2021 • Daniel Barath, Dmytro Mishkin, Ivan Eichhardt, Ilia Shipachev, Jiri Matas
We propose ways to speed up the initial pose-graph generation for global Structure-from-Motion algorithms.
no code implementations • 9 Nov 2020 • Yanlin Qian, Miaojing Shi, Joni-Kristian Kämäräinen, Jiri Matas
We address the problem of decomposing an image into albedo and shading.
no code implementations • 11 Oct 2020 • Dmytro Mishkin, Amy Tabb, Jiri Matas
We claim, and present evidence, that allowing arXiv publication before a conference or journal submission benefits researchers, especially early career, as well as the whole scientific community.
4 code implementations • 15 Sep 2020 • Tomas Hodan, Martin Sundermeyer, Bertram Drost, Yann Labbe, Eric Brachmann, Frank Michel, Carsten Rother, Jiri Matas
This paper presents the evaluation methodology, datasets, and results of the BOP Challenge 2020, the third in a series of public competitions organized with the goal to capture the status quo in the field of 6D object pose estimation from an RGB-D image.
no code implementations • 6 Jul 2020 • Klára Janoušková, Jiri Matas, Lluis Gomez, Dimosthenis Karatzas
We present a method for exploiting weakly annotated images to improve text extraction pipelines.
no code implementations • ECCV 2020 • Yash Patel, Tomas Hodan, Jiri Matas
The effectiveness of the proposed technique is demonstrated in a post-tuning setup, where a trained model is tuned using the learned surrogate.
2 code implementations • ECCV 2020 • Lam Huynh, Phong Nguyen-Ha, Jiri Matas, Esa Rahtu, Janne Heikkila
Recovering the scene depth from a single image is an ill-posed problem that requires additional priors, often referred to as monocular depth cues, to disambiguate different 3D interpretations.
1 code implementation • CVPR 2020 • Tomas Hodan, Daniel Barath, Jiri Matas
A data-dependent number of corresponding 3D locations is selected per pixel, and poses of possibly multiple object instances are estimated using a robust and efficient variant of the PnP-RANSAC algorithm.
3 code implementations • 8 Mar 2020 • Yanlin Qian, Jani Käpylä, Joni-Kristian Kämäräinen, Samu Koskinen, Jiri Matas
The conventional approach is to use a single frame - shot frame - to estimate the scene illumination color.
5 code implementations • 3 Mar 2020 • Yuhe Jin, Dmytro Mishkin, Anastasiia Mishchuk, Jiri Matas, Pascal Fua, Kwang Moo Yi, Eduard Trulls
We introduce a comprehensive benchmark for local features and robust estimation algorithms, focusing on the downstream task -- the accuracy of the reconstructed camera pose -- as our primary metric.
no code implementations • CVPR 2020 • Daniel Barath, Jana Noskova, Maksym Ivashechkin, Jiri Matas
A new method for robust estimation, MAGSAC++, is proposed.
no code implementations • 2 Dec 2019 • Yanlin Qian, Alan Lukežič, Matej Kristan, Joni-Kristian Kämäräinen, Jiri Matas
In this work, we propose a deep depth-aware long-term tracker that achieves state-of-the-art RGBD tracking performance and is fast to run.
2 code implementations • CVPR 2020 • Denys Rozumnyi, Jan Kotera, Filip Sroubek, Jiri Matas
We propose a novel method that tracks fast moving objects, mainly non-uniform spherical, in full 6 degrees of freedom, estimating simultaneously their 3D motion trajectory, 3D pose and object appearance changes with a time step that is a fraction of the video frame exposure time.
Ranked #1 on Video Super-Resolution on TbD
no code implementations • 1 Jul 2019 • Nibal Nayef, Yash Patel, Michal Busta, Pinaki Nath Chowdhury, Dimosthenis Karatzas, Wafa Khlif, Jiri Matas, Umapada Pal, Jean-Christophe Burie, Cheng-Lin Liu, Jean-Marc Ogier
With the growing cosmopolitan culture of modern cities, the need of robust Multi-Lingual scene Text (MLT) detection and recognition systems has never been more immense.
Cultural Vocal Bursts Intensity Prediction General Classification +1
2 code implementations • ICCV 2019 • Daniel Barath, Jiri Matas
The Progressive-X algorithm, Prog-X in short, is proposed for geometric multi-model fitting.
no code implementations • 5 Jun 2019 • Daniel Barath, Maksym Ivashechkin, Jiri Matas
We propose Progressive NAPSAC, P-NAPSAC in short, which merges the advantages of local and global sampling by drawing samples from gradually growing neighborhoods.
1 code implementation • 28 Feb 2019 • Matej Smid, Jiri Matas
A simple method for synchronization of video streams with a precision better than one millisecond is proposed.
no code implementations • 27 Feb 2019 • Yanlin Qian, Song Yan, Joni-Kristian Kämäräinen, Jiri Matas
In the real world, a scene is usually cast by multiple illuminants and herein we address the problem of spatial illumination estimation.
2 code implementations • CVPR 2019 • Yanlin Qian, Joni-Kristian Kämäräinen, Jarno Nikkanen, Jiri Matas
We propose a novel grayness index for finding gray pixels and demonstrate its effectiveness and efficiency in illumination estimation.
no code implementations • CVPR 2019 • Ugur Kart, Alan Lukezic, Matej Kristan, Joni-Kristian Kamarainen, Jiri Matas
Standard RGB-D trackers treat the target as an inherently 2D structure, which makes modelling appearance changes related even to simple out-of-plane rotation highly challenging.
no code implementations • 23 Nov 2018 • Janne Mustaniemi, Juho Kannala, Jiri Matas, Simo Särkkä, Janne Heikkilä
The paper addresses the problem of acquiring high-quality photographs with handheld smartphone cameras in low-light imaging conditions.
no code implementations • 9 Oct 2018 • Tomas Hodan, Rigas Kouskouridas, Tae-Kyun Kim, Federico Tombari, Kostas Bekris, Bertram Drost, Thibault Groueix, Krzysztof Walas, Vincent Lepetit, Ales Leonardis, Carsten Steger, Frank Michel, Caner Sahin, Carsten Rother, Jiri Matas
The workshop featured four invited talks, oral and poster presentations of accepted workshop papers, and an introduction of the BOP benchmark for 6D object pose estimation.
1 code implementation • 1 Oct 2018 • Janne Mustaniemi, Juho Kannala, Simo Särkkä, Jiri Matas, Janne Heikkilä
We propose a deblurring method that incorporates gyroscope measurements into a convolutional neural network (CNN).
1 code implementation • ECCV 2018 • Tomas Hodan, Frank Michel, Eric Brachmann, Wadim Kehl, Anders Glent Buch, Dirk Kraft, Bertram Drost, Joel Vidal, Stephan Ihrke, Xenophon Zabulis, Caner Sahin, Fabian Manhardt, Federico Tombari, Tae-Kyun Kim, Jiri Matas, Carsten Rother
We propose a benchmark for 6D pose estimation of a rigid object from a single RGB-D input image.
no code implementations • 17 Aug 2018 • Junsheng Fu, Said Pertuz, Jiri Matas, Joni-Kristian Kämäräinen
We consider a single-query 6-DoF camera pose estimation with reference images and a point cloud, i. e. the problem of estimating the position and orientation of a camera by using reference images and a point cloud.
no code implementations • 22 May 2018 • Janne Mustaniemi, Juho Kannala, Simo Särkkä, Jiri Matas, Janne Heikkilä
It is well-known that motion blur decreases the performance of traditional feature detectors and descriptors.
no code implementations • 21 May 2018 • Milan Sulc, Jiri Matas
The proposed Maximum a Posteriori estimation increases the prediction accuracy by 2. 8% on PlantCLEF 2017 and by 1. 8% on FGVCx Fungi, where the existing MLE method would lead to a decrease accuracy.
1 code implementation • 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018) 2018 • Pavel Jahoda, Antonin Vobecky, Jan Cech, Jiri Matas
We propose a baseline method combining a deep convolutional neural network with an SVM.
1 code implementation • 22 Mar 2018 • Yanlin Qian, Said Pertuz, Jarno Nikkanen, Joni-Kristian Kämäräinen, Jiri Matas
We present a statistical color constancy method that relies on novel gray pixel detection and mean shift clustering.
2 code implementations • CVPR 2019 • Daniel Barath, Jana Noskova, Jiri Matas
A method called, sigma-consensus, is proposed to eliminate the need for a user-defined inlier-outlier threshold in RANSAC.
3 code implementations • 30 Jan 2018 • Michal Bušta, Yash Patel, Jiri Matas
An end-to-end trainable (fully differentiable) method for multi-language scene text localization and recognition is proposed.
12 code implementations • CVPR 2018 • Orest Kupyn, Volodymyr Budzan, Mykola Mykhailych, Dmytro Mishkin, Jiri Matas
The quality of the deblurring model is also evaluated in a novel way on a real-world problem -- object detection on (de-)blurred images.
Ranked #3 on Deblurring on REDS
3 code implementations • ECCV 2018 • Dmytro Mishkin, Filip Radenovic, Jiri Matas
A method for learning local affine-covariant regions is presented.
Ranked #4 on Image Matching on IMC PhotoTourism (using extra training data)
no code implementations • ICCV 2017 • Yanlin Qian, Ke Chen, Jarno Nikkanen, Joni-Kristian Kamarainen, Jiri Matas
We introduce a novel formulation of temporal color constancy which considers multiple frames preceding the frame for which illumination is estimated.
no code implementations • ICCV 2017 • Michal Busta, Lukas Neumann, Jiri Matas
A method for scene text localization and recognition is proposed.
1 code implementation • CVPR 2018 • Daniel Barath, Jiri Matas
A novel method for robust estimation, called Graph-Cut RANSAC, GC-RANSAC in short, is introduced.
1 code implementation • ECCV 2018 • Daniel Barath, Jiri Matas
The move replaces a set of labels with the corresponding density mode in the model parameter domain, thus achieving fast and robust optimization.
4 code implementations • NeurIPS 2017 • Anastasiya Mishchuk, Dmytro Mishkin, Filip Radenovic, Jiri Matas
We introduce a novel loss for learning local feature descriptors which is inspired by the Lowe's matching criterion for SIFT.
2 code implementations • 19 Jan 2017 • Tomas Hodan, Pavel Haluza, Stepan Obdrzalek, Jiri Matas, Manolis Lourakis, Xenophon Zabulis
There are approximately 39K training and 10K test images from each sensor.
1 code implementation • 29 Nov 2016 • Janne Mustaniemi, Juho Kannala, Simo Särkkä, Jiri Matas, Janne Heikkilä
In the process, we also perform a temporal and spatial alignment of the camera and the IMU.
3 code implementations • CVPR 2017 • Denys Rozumnyi, Jan Kotera, Filip Sroubek, Lukas Novotny, Jiri Matas
The notion of a Fast Moving Object (FMO), i. e. an object that moves over a distance exceeding its size within the exposure time, is introduced.
1 code implementation • 24 Aug 2016 • Javier Aldana-Iuit, Dmytro Mishkin, Ondrej Chum, Jiri Matas
A novel similarity-covariant feature detector that extracts points whose neighbourhoods, when treated as a 3D intensity surface, have a saddle-like intensity profile.
no code implementations • 13 Jul 2016 • Yanlin Qian, Ke Chen, Joni-Kristian Kamarainen, Jarno Nikkanen, Jiri Matas
Computational color constancy that requires esti- mation of illuminant colors of images is a fundamental yet active problem in computer vision, which can be formulated into a regression problem.
1 code implementation • 7 Jun 2016 • Dmytro Mishkin, Nikolay Sergievskiy, Jiri Matas
The paper systematically studies the impact of a range of recent advances in CNN architectures and learning methods on the object categorization (ILSVRC) problem.
no code implementations • CVPR 2016 • Filip Radenovic, Johannes L. Schonberger, Dinghuang Ji, Jan-Michael Frahm, Ondrej Chum, Jiri Matas
We present an algorithm that leverages the appearance variety to obtain more complete and accurate scene geometry along with consistent multi-illumination appearance information.
4 code implementations • 26 Jan 2016 • Andreas Veit, Tomas Matera, Lukas Neumann, Jiri Matas, Serge Belongie
The goal of COCO-Text is to advance state-of-the-art in text detection and recognition in natural images.
no code implementations • ICCV 2015 • Michal Busta, Lukas Neumann, Jiri Matas
After a novel efficient classification step, the number of regions is reduced to 7 times less than the standard method and is still almost 3 times faster.
7 code implementations • ICLR 2015 • Dmytro Mishkin, Jiri Matas
Experiment with different activation functions (maxout, ReLU-family, tanh) show that the proposed initialization leads to learning of very deep nets that (i) produces networks with test accuracy better or equal to standard methods and (ii) is at least as fast as the complex schemes proposed specifically for very deep nets such as FitNets (Romero et al. (2015)) and Highway (Srivastava et al. (2015)).
Ranked #23 on Image Classification on MNIST
no code implementations • ICCV 2015 • David Novotny, Jiri Matas
The efficiency is achieved by the use of spatial bins in a novel combination with sparsity-inducing group normalized SVM.
2 code implementations • 24 Apr 2015 • Dmytro Mishkin, Jiri Matas, Michal Perdoch, Karel Lenc
We have presented a new problem -- the wide multiple baseline stereo (WxBS) -- which considers matching of images that simultaneously differ in more than one image acquisition factor such as viewpoint, illumination, sensor type or where object appearance changes significantly, e. g. over time.
no code implementations • 23 Apr 2015 • Tomas Vojir, Jiri Matas, Jana Noskova
We show the effectiveness of the proposed method on combination of two and three tracking algorithms.
2 code implementations • 9 Mar 2015 • Dmytro Mishkin, Jiri Matas, Michal Perdoch
A novel algorithm for wide-baseline matching called MODS - Matching On Demand with view Synthesis - is presented.
no code implementations • 4 Mar 2015 • Matej Kristan, Jiri Matas, Ales Leonardis, Tomas Vojir, Roman Pflugfelder, Gustavo Fernandez, Georg Nebehay, Fatih Porikli, Luka Cehovin
This paper addresses the problem of single-target tracker performance evaluation.
no code implementations • CVPR 2014 • James Pritts, Ondrej Chum, Jiri Matas
This paper presents a novel and general method for the detection, rectification and segmentation of imaged coplanar repeated patterns.
no code implementations • 24 Nov 2013 • Patrick Ott, Mark Everingham, Jiri Matas
An "elephant in the room" for most current object detection and localization methods is the lack of explicit modelling of partial visibility due to occlusion by other objects or truncation by the image boundary.
no code implementations • 17 Jun 2013 • Dmytro Mishkin, Michal Perdoch, Jiri Matas
Wide-baseline matching focussing on problems with extreme viewpoint change is considered.