Search Results for author: Luc van Gool

Found 335 papers, 149 papers with code

Fixing Localization Errors to Improve Image Classification

1 code implementation ECCV 2020 Guolei Sun, Salman Khan, Wen Li, Hisham Cholakkal, Fahad Shahbaz Khan, Luc van Gool

This way, in an effort to fix localization errors, our loss provides an extra supervisory signal that helps the model to better discriminate between similar classes.

Classification General Classification +3

Modeling the Effects of Windshield Refraction for Camera Calibration

no code implementations ECCV 2020 Frank Verbiest, Marc Proesmans, Luc van Gool

Instead of using a generalized camera approach, we propose a novel approach to jointly optimize a traditional camera model, and a mathematical representation of the windshield’s surface.

Autonomous Driving

3D Compositional Zero-shot Learning with DeCompositional Consensus

no code implementations29 Nov 2021 Muhammad Ferjad Naeem, Evin Pınar Örnek, Yongqin Xian, Luc van Gool, Federico Tombari

Parts represent a basic unit of geometric and semantic similarity across different objects.

Predict, Prevent, and Evaluate: Disentangled Text-Driven Image Manipulation Empowered by Pre-Trained Vision-Language Model

no code implementations26 Nov 2021 Zipeng Xu, Tianwei Lin, Hao Tang, Fu Li, Dongliang He, Nicu Sebe, Radu Timofte, Luc van Gool, Errui Ding

In this paper, we propose a novel framework, i. e., Predict, Prevent, and Evaluate (PPE), for disentangled text-driven image manipulation, which does not need manual annotation and thus is not limited to fixed manipulations.

Image Manipulation Language Modelling

MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimation

1 code implementation24 Nov 2021 Wenhao Li, Hong Liu, Hao Tang, Pichao Wang, Luc van Gool

Estimating 3D human poses from monocular videos is a challenging task due to depth ambiguity and self-occlusion.

3D Human Pose Estimation

Mask-guided Spectral-wise Transformer for Efficient Hyperspectral Image Reconstruction

no code implementations15 Nov 2021 Yuanhao Cai, Jing Lin, Xiaowan Hu, Haoqian Wang, Xin Yuan, Yulun Zhang, Radu Timofte, Luc van Gool

Hyperspectral image (HSI) reconstruction aims to recover the 3D spatial-spectral signal from a 2D measurement in the coded aperture snapshot spectral imaging (CASSI) system.

Image Reconstruction

Normalizing Flow as a Flexible Fidelity Objective for Photo-Realistic Super-resolution

no code implementations5 Nov 2021 Andreas Lugmayr, Martin Danelljan, Fisher Yu, Luc van Gool, Radu Timofte

Super-resolution is an ill-posed problem, where a ground-truth high-resolution image represents only one possibility in the space of plausible solutions.


Neural Radiance Fields Approach to Deep Multi-View Photometric Stereo

no code implementations11 Oct 2021 Berk Kaya, Suryansh Kumar, Francesco Sarno, Vittorio Ferrari, Luc van Gool

Our method performs neural rendering of multi-view images while utilizing surface normals estimated by a deep photometric stereo network.

3D Reconstruction Neural Rendering

Neural Architecture Search for Efficient Uncalibrated Deep Photometric Stereo

no code implementations11 Oct 2021 Francesco Sarno, Suryansh Kumar, Berk Kaya, Zhiwu Huang, Vittorio Ferrari, Luc van Gool

We then perform a continuous relaxation of this search space and present a gradient-based optimization strategy to find an efficient light calibration and normal estimation network.

Neural Architecture Search

Structured Bird's-Eye-View Traffic Scene Understanding from Onboard Images

1 code implementation ICCV 2021 Yigit Baran Can, Alexander Liniger, Danda Pani Paudel, Luc van Gool

In this work, we study the problem of extracting a directed graph representing the local road network in BEV coordinates, from a single onboard camera image.

Autonomous Navigation Scene Understanding

PDC-Net+: Enhanced Probabilistic Dense Correspondence Network

1 code implementation28 Sep 2021 Prune Truong, Martin Danelljan, Radu Timofte, Luc van Gool

In order to apply dense methods to real-world applications, such as pose estimation, image manipulation, or 3D reconstruction, it is therefore crucial to estimate the confidence of the predicted matches.

3D Reconstruction Geometric Matching +5

Context-aware Padding for Semantic Segmentation

no code implementations16 Sep 2021 Yu-Hui Huang, Marc Proesmans, Luc van Gool

Zero padding is widely used in convolutional neural networks to prevent the size of feature maps diminishing too fast.

Semantic Segmentation

TADA: Taxonomy Adaptive Domain Adaptation

no code implementations10 Sep 2021 Rui Gong, Martin Danelljan, Dengxin Dai, Wenguan Wang, Danda Pani Paudel, Ajad Chhatkuli, Fisher Yu, Luc van Gool

We extensively evaluate the effectiveness of our framework under different TADA settings: open taxonomy, coarse-to-fine taxonomy, and partially-overlapping taxonomy.

Contrastive Learning Domain Adaptation

Perceptual Learned Video Compression with Recurrent Conditional GAN

1 code implementation7 Sep 2021 Ren Yang, Luc van Gool, Radu Timofte

The user study further validates the outstanding perceptual performance of PLVC in comparison with the latest learned video compression approaches and the official HEVC test model (HM 16. 20).

Video Compression

Improving Semi-Supervised and Domain-Adaptive Semantic Segmentation with Self-Supervised Depth Estimation

1 code implementation28 Aug 2021 Lukas Hoyer, Dengxin Dai, Qin Wang, Yuhua Chen, Luc van Gool

Training deep networks for semantic segmentation requires large amounts of labeled training data, which presents a major challenge in practice, as labeling segmentation masks is a highly labor-intensive process.

Data Augmentation Domain Adaptation +3

Generalized Real-World Super-Resolution through Adversarial Robustness

1 code implementation25 Aug 2021 Angela Castillo, María Escobar, Juan C. Pérez, Andrés Romero, Radu Timofte, Luc van Gool, Pablo Arbeláez

Instead of learning a dataset-specific degradation, we employ adversarial attacks to create difficult examples that target the model's weaknesses.

Adversarial Robustness Super-Resolution

SwinIR: Image Restoration Using Swin Transformer

2 code implementations23 Aug 2021 Jingyun Liang, JieZhang Cao, Guolei Sun, Kai Zhang, Luc van Gool, Radu Timofte

In particular, the deep feature extraction module is composed of several residual Swin Transformer blocks (RSTB), each of which has several Swin Transformer layers together with a residual connection.

Color Image Denoising Image Denoising +4

End-to-End Urban Driving by Imitating a Reinforcement Learning Coach

1 code implementation ICCV 2021 Zhejun Zhang, Alexander Liniger, Dengxin Dai, Fisher Yu, Luc van Gool

Our end-to-end agent achieves a 78% success rate while generalizing to a new town and new weather on the NoCrash-dense benchmark and state-of-the-art performance on the challenging public routes of the CARLA LeaderBoard.

Autonomous Driving Imitation Learning

Deep Reparametrization of Multi-Frame Super-Resolution and Denoising

2 code implementations ICCV 2021 Goutam Bhat, Martin Danelljan, Fisher Yu, Luc van Gool, Radu Timofte

The deep reparametrization allows us to directly model the image formation process in the latent space, and to integrate learned image priors into the prediction.

Denoising Image Restoration +1

Decoder Fusion RNN: Context and Interaction Aware Decoders for Trajectory Prediction

no code implementations12 Aug 2021 Edoardo Mello Rella, Jan-Nico Zaech, Alexander Liniger, Luc van Gool

Forecasting the future behavior of all traffic agents in the vicinity is a key task to achieve safe and reliable autonomous driving systems.

Motion Forecasting Trajectory Prediction

Mutual Affine Network for Spatially Variant Kernel Estimation in Blind Image Super-Resolution

1 code implementation ICCV 2021 Jingyun Liang, Guolei Sun, Kai Zhang, Luc van Gool, Radu Timofte

Extensive experiments on synthetic and real images show that the proposed MANet not only performs favorably for both spatially variant and invariant kernel estimation, but also leads to state-of-the-art blind SR performance when combined with non-blind SR methods.

Affine Transformation Image Super-Resolution

Fog Simulation on Real LiDAR Point Clouds for 3D Object Detection in Adverse Weather

1 code implementation ICCV 2021 Martin Hahner, Christos Sakaridis, Dengxin Dai, Luc van Gool

2) Through extensive experiments with several state-of-the-art detection approaches, we show that our fog simulation can be leveraged to significantly improve the performance for 3D object detection in the presence of fog.

3D Object Detection Physical Simulations

Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescaling

1 code implementation ICCV 2021 Jingyun Liang, Andreas Lugmayr, Kai Zhang, Martin Danelljan, Luc van Gool, Radu Timofte

More specifically, HCFlow learns a bijective mapping between HR and LR image pairs by modelling the distribution of the LR image and the rest high-frequency component simultaneously.

Image Super-Resolution

Boosting Few-shot Semantic Segmentation with Transformers

no code implementations4 Aug 2021 Guolei Sun, Yun Liu, Jingyun Liang, Luc van Gool

Due to the fact that fully supervised semantic segmentation methods require sufficient fully-labeled data to work well and can not generalize to unseen classes, few-shot segmentation has attracted lots of research attention.

Few-Shot Semantic Segmentation Semantic Segmentation

A Survey on Deep Learning Technique for Video Segmentation

no code implementations2 Jul 2021 Wenguan Wang, Tianfei Zhou, Fatih Porikli, David Crandall, Luc van Gool

Video segmentation, i. e., partitioning video frames into multiple segments or objects, plays a critical role in a broad range of practical applications, e. g., visual effect assistance in movie, scene understanding in autonomous driving, and virtual background creation in video conferencing, to name a few.

Autonomous Driving Scene Understanding +3

On the Practicality of Deterministic Epistemic Uncertainty

1 code implementation1 Jul 2021 Janis Postels, Mattia Segu, Tao Sun, Luc van Gool, Fisher Yu, Federico Tombari

A set of novel approaches for estimating epistemic uncertainty in deep neural networks with a single forward pass has recently emerged as a valid alternative to Bayesian Neural Networks.

Image Classification Semantic Segmentation

GANmut: Learning Interpretable Conditional Space for Gamut of Emotions

no code implementations CVPR 2021 Stefano d'Apolito, Danda Pani Paudel, Zhiwu Huang, Andres Romero, Luc van Gool

On the other hand, learning from inexpensive and intuitive basic categorical emotion labels leads to limited emotion variability.

Video Super-Resolution Transformer

1 code implementation12 Jun 2021 JieZhang Cao, Yawei Li, Kai Zhang, Luc van Gool

Specifically, to tackle the first issue, we present a spatial-temporal convolutional self-attention layer with a theoretical understanding to exploit the locality information.

Optical Flow Estimation Video Super-Resolution

Generative Flows with Invertible Attentions

no code implementations7 Jun 2021 Rhea Sanjay Sukthanker, Zhiwu Huang, Suryansh Kumar, Radu Timofte, Luc van Gool

Flow-based generative models have shown excellent ability to explicitly learn the probability density function of data via a sequence of invertible transformations.

Transformer in Convolutional Neural Networks

1 code implementation6 Jun 2021 Yun Liu, Guolei Sun, Yu Qiu, Le Zhang, Ajad Chhatkuli, Luc van Gool

We tackle the low-efficiency flaw of vision transformer caused by the high computational/space complexity in Multi-Head Self-Attention (MHSA).

Go with the Flows: Mixtures of Normalizing Flows for Point Cloud Generation and Reconstruction

no code implementations6 Jun 2021 Janis Postels, Mengya Liu, Riccardo Spezialetti, Luc van Gool, Federico Tombari

Recently normalizing flows (NFs) have demonstrated state-of-the-art performance on modeling 3D point clouds while allowing sampling with arbitrary resolution at inference time.

Data Augmentation Point Cloud Generation

Fourier Space Losses for Efficient Perceptual Image Super-Resolution

no code implementations ICCV 2021 Dario Fuoli, Luc van Gool, Radu Timofte

As large models are often not practical in real-world applications, we investigate and propose novel loss functions, to enable SR with high perceptual quality from much more efficient models.

Image Super-Resolution

Boosting Crowd Counting with Transformers

no code implementations23 May 2021 Guolei Sun, Yun Liu, Thomas Probst, Danda Pani Paudel, Nikola Popovic, Luc van Gool

This indicates that global scene context is essential, despite the seemingly bottom-up nature of the problem.

Crowd Counting

Learning to Relate Depth and Semantics for Unsupervised Domain Adaptation

1 code implementation CVPR 2021 Suman Saha, Anton Obukhov, Danda Pani Paudel, Menelaos Kanakis, Yuhua Chen, Stamatios Georgoulis, Luc van Gool

Specifically, we show that: (1) our approach improves performance on all tasks when they are complementary and mutually dependent; (2) the CTRL helps to improve both semantic segmentation and depth estimation tasks performance in the challenging UDA setting; (3) the proposed ISL training scheme further improves the semantic segmentation performance.

Monocular Depth Estimation Multi-Task Learning +2

Exploring Relational Context for Multi-Task Dense Prediction

1 code implementation ICCV 2021 David Bruggemann, Menelaos Kanakis, Anton Obukhov, Stamatios Georgoulis, Luc van Gool

Our goal is to find the most efficient way to refine each task prediction by capturing cross-task contexts dependent on tasks' relations.

Neural Architecture Search

ACDC: The Adverse Conditions Dataset with Correspondences for Semantic Driving Scene Understanding

no code implementations ICCV 2021 Christos Sakaridis, Dengxin Dai, Luc van Gool

To address this, we introduce ACDC, the Adverse Conditions Dataset with Correspondences for training and testing semantic segmentation methods on adverse visual conditions.

Scene Understanding Self-Driving Cars +1

Learnable Online Graph Representations for 3D Multi-Object Tracking

no code implementations23 Apr 2021 Jan-Nico Zaech, Dengxin Dai, Alexander Liniger, Martin Danelljan, Luc van Gool

Tracking of objects in 3D is a fundamental task in computer vision that finds use in a wide range of applications such as autonomous driving, robotics or augmented reality.

3D Multi-Object Tracking Autonomous Driving

Towards Efficient Graph Convolutional Networks for Point Cloud Handling

no code implementations ICCV 2021 Yawei Li, He Chen, Zhaopeng Cui, Radu Timofte, Marc Pollefeys, Gregory Chirikjian, Luc van Gool

In this paper, we aim at improving the computational efficiency of graph convolutional networks (GCNs) for learning on point clouds.

LocalViT: Bringing Locality to Vision Transformers

2 code implementations12 Apr 2021 Yawei Li, Kai Zhang, JieZhang Cao, Radu Timofte, Luc van Gool

The importance of locality mechanisms is validated in two ways: 1) A wide range of design choices (activation function, layer placement, expansion ratio) are available for incorporating locality mechanisms and all proper choices can lead to a performance gain over the baseline, and 2) The same locality mechanism is successfully applied to 4 vision transformers, which shows the generalization of the locality concept.

Image Classification

Warp Consistency for Unsupervised Learning of Dense Correspondences

2 code implementations ICCV 2021 Prune Truong, Martin Danelljan, Fisher Yu, Luc van Gool

From our observations and empirical results, we design a general unsupervised objective employing two of the derived constraints.

Dense Pixel Correspondence Estimation

Learning Target Candidate Association to Keep Track of What Not to Track

1 code implementation ICCV 2021 Christoph Mayer, Martin Danelljan, Danda Pani Paudel, Luc van Gool

To tackle the problem of lacking ground-truth correspondences between distractor objects in visual tracking, we propose a training strategy that combines partial annotations with self-supervision.

Visual Object Tracking Visual Tracking

Unsupervised Robust Domain Adaptation without Source Data

no code implementations26 Mar 2021 Peshal Agarwal, Danda Pani Paudel, Jan-Nico Zaech, Luc van Gool

This paper aims at answering the question of finding the right strategy to make the target model robust and accurate in the setting of unsupervised domain adaptation without source data.

Image Classification Unsupervised Domain Adaptation

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution

1 code implementation ICCV 2021 Kai Zhang, Jingyun Liang, Luc van Gool, Radu Timofte

It is widely acknowledged that single image super-resolution (SISR) methods would not perform well if the assumed degradation model deviates from those in real images.

Image Super-Resolution

Temporally-Weighted Hierarchical Clustering for Unsupervised Action Segmentation

2 code implementations CVPR 2021 M. Saquib Sarfraz, Naila Murray, Vivek Sharma, Ali Diba, Luc van Gool, Rainer Stiefelhagen

Action segmentation refers to inferring boundaries of semantically consistent visual concepts in videos and is an important requirement for many video understanding tasks.

Action Segmentation Video Understanding

Differentiable Multi-Granularity Human Representation Learning for Instance-Aware Human Semantic Parsing

1 code implementation CVPR 2021 Tianfei Zhou, Wenguan Wang, Si Liu, Yi Yang, Luc van Gool

To address the challenging task of instance-aware human part parsing, a new bottom-up regime is proposed to learn category-level human semantic segmentation as well as multi-person pose estimation in a joint and end-to-end manner.

Human Parsing Multi-Person Pose Estimation +3

Exploring Cross-Image Pixel Contrast for Semantic Segmentation

3 code implementations ICCV 2021 Wenguan Wang, Tianfei Zhou, Fisher Yu, Jifeng Dai, Ender Konukoglu, Luc van Gool

Inspired by the recent advance in unsupervised contrastive representation learning, we propose a pixel-wise contrastive framework for semantic segmentation in the fully supervised setting.

Metric Learning Optical Character Recognition +2

Hyperspectral Image Super-Resolution with Spectral Mixup and Heterogeneous Datasets

2 code implementations19 Jan 2021 Ke Li, Dengxin Dai, Ender Konukoglu, Luc van Gool

With these contributions, our method is able to learn from heterogeneous datasets and lift the requirement for having a large amount of HD HSI training samples.

Data Augmentation Hyperspectral Image Super-Resolution +1

Trilevel Neural Architecture Search for Efficient Single Image Super-Resolution

no code implementations17 Jan 2021 Yan Wu, Zhiwu Huang, Suryansh Kumar, Rhea Sanjay Sukthanker, Radu Timofte, Luc van Gool

Modern solutions to the single image super-resolution (SISR) problem using deep neural networks aim not only at better performance accuracy but also at a lighter and computationally efficient model.

Image Super-Resolution Neural Architecture Search

Local Memory Attention for Fast Video Semantic Segmentation

1 code implementation5 Jan 2021 Matthieu Paul, Martin Danelljan, Luc van Gool, Radu Timofte

Our approach aggregates a rich representation of the semantic information in past frames into a memory module.

Semantic Segmentation Video Semantic Segmentation

The Card Shuffling Hypotheses: Building a Time and Memory Efficient Graph Convolutional Network

no code implementations1 Jan 2021 Yawei Li, He Chen, Zhaopeng Cui, Radu Timofte, Marc Pollefeys, Gregory Chirikjian, Luc van Gool

State-of-the-art GCNs adopt $K$-nearest neighbor (KNN) searches for local feature aggregation and feature extraction operations from layer to layer.

3D Classification Point Cloud Classification +1

Vi2CLR: Video and Image for Visual Contrastive Learning of Representation

no code implementations ICCV 2021 Ali Diba, Vivek Sharma, Reza Safdari, Dariush Lotfi, Saquib Sarfraz, Rainer Stiefelhagen, Luc van Gool

In this paper, we introduce a novel self-supervised visual representation learning method which understands both images and videos in a joint learning fashion.

Action Recognition Contrastive Learning +2

Exploring Geometry-Aware Contrast and Clustering Harmonization for Self-Supervised 3D Object Detection

no code implementations ICCV 2021 Hanxue Liang, Chenhan Jiang, Dapeng Feng, Xin Chen, Hang Xu, Xiaodan Liang, Wei zhang, Zhenguo Li, Luc van Gool

Here we present a novel self-supervised 3D Object detection framework that seamlessly integrates the geometry-aware contrast and clustering harmonization to lift the unsupervised 3D representation learning, named GCC-3D.

3D Object Detection Representation Learning +1

Unsupervised Monocular Depth Reconstruction of Non-Rigid Scenes

no code implementations31 Dec 2020 Ayça Takmaz, Danda Pani Paudel, Thomas Probst, Ajad Chhatkuli, Martin R. Oswald, Luc van Gool

In this work, we present an unsupervised monocular framework for dense depth estimation of dynamic scenes, which jointly reconstructs rigid and non-rigid parts without explicitly modelling the camera motion.

Depth Estimation Motion Segmentation

An Efficient Recurrent Adversarial Framework for Unsupervised Real-Time Video Enhancement

no code implementations24 Dec 2020 Dario Fuoli, Zhiwu Huang, Danda Pani Paudel, Luc van Gool, Radu Timofte

Video enhancement is a challenging problem, more than that of stills, mainly due to high computational cost, larger data volumes and the difficulty of achieving consistency in the spatio-temporal domain.

Video Enhancement

Three Ways to Improve Semantic Segmentation with Self-Supervised Depth Estimation

1 code implementation CVPR 2021 Lukas Hoyer, Dengxin Dai, Yuhua Chen, Adrian Köring, Suman Saha, Luc van Gool

Training deep networks for semantic segmentation requires large amounts of labeled training data, which presents a major challenge in practice, as labeling segmentation masks is a highly labor-intensive process.

Data Augmentation Monocular Depth Estimation +1

CompositeTasking: Understanding Images by Spatial Composition of Tasks

1 code implementation CVPR 2021 Nikola Popovic, Danda Pani Paudel, Thomas Probst, Guolei Sun, Luc van Gool

Learning to perform spatially distributed tasks is motivated by the frequent availability of only sparse labels across tasks, and the desire for a compact multi-tasking network.

Cluster, Split, Fuse, and Update: Meta-Learning for Open Compound Domain Adaptive Semantic Segmentation

no code implementations CVPR 2021 Rui Gong, Yuhua Chen, Danda Pani Paudel, Yawei Li, Ajad Chhatkuli, Wen Li, Dengxin Dai, Luc van Gool

Open compound domain adaptation (OCDA) is a domain adaptation setting, where target domain is modeled as a compound of multiple unknown homogeneous domains, which brings the advantage of improved generalization to unseen domains.

Domain Adaptation Meta-Learning +2

Scaling Semantic Segmentation Beyond 1K Classes on a Single GPU

1 code implementation ICCV 2021 Shipra Jain, Danda Paudel Pani, Martin Danelljan, Luc van Gool

In this paper, we propose a novel training methodology to train and scale the existing semantic segmentation models for a large number of semantic classes without increasing the memory overhead.

Image Classification Object Detection +1

Understanding Bird's-Eye View Semantic HD-Maps Using an Onboard Monocular Camera

no code implementations5 Dec 2020 Yigit Baran Can, Alexander Liniger, Ozan Unal, Danda Paudel, Luc van Gool

We study three key aspects of this task, image-level understanding, BEV level understanding, and the aggregation of temporal information.

Autonomous Navigation Scene Understanding

The Hidden Uncertainty in a Neural Networks Activations

no code implementations5 Dec 2020 Janis Postels, Hermann Blum, Yannick Strümpler, Cesar Cadena, Roland Siegwart, Luc van Gool, Federico Tombari

We find that this leads to improved OOD detection of epistemic uncertainty at the cost of ambiguous calibration close to the data distribution.

Density Estimation

Learning from Simulation, Racing in Reality

no code implementations26 Nov 2020 Eugenio Chisari, Alexander Liniger, Alisa Rupenyan, Luc van Gool, John Lygeros

We present a reinforcement learning-based solution to autonomously race on a miniature race car platform.

3D CNNs with Adaptive Temporal Feature Resolutions

1 code implementation CVPR 2021 Mohsen Fayyaz, Emad Bahrami, Ali Diba, Mehdi Noroozi, Ehsan Adeli, Luc van Gool, Juergen Gall

While the GFLOPs of a 3D CNN can be decreased by reducing the temporal feature resolution within the network, there is no setting that is optimal for all input clips.

Action Recognition

Zero-Pair Image to Image Translation using Domain Conditional Normalization

1 code implementation11 Nov 2020 Samarth Shukla, Andrés Romero, Luc van Gool, Radu Timofte

In this paper, we propose an approach based on domain conditional normalization (DCN) for zero-pair image-to-image translation, i. e., translating between two domains which have no paired training data available but each have paired training data with a third domain.

Image-to-Image Translation Translation

Neural Architecture Search of SPD Manifold Networks

1 code implementation27 Oct 2020 Rhea Sanjay Sukthanker, Zhiwu Huang, Suryansh Kumar, Erik Goron Endsjo, Yan Wu, Luc van Gool

To address this problem, we first introduce a geometrically rich and diverse SPD neural architecture search space for an efficient SPD cell design.

Emotion Recognition Neural Architecture Search

Self-Supervised Shadow Removal

no code implementations22 Oct 2020 Florin-Alexandru Vasluianu, Andres Romero, Luc van Gool, Radu Timofte

Shadow removal is an important computer vision task aiming at the detection and successful removal of the shadow produced by an occluded light source and a photo-realistic restoration of the image contents.

Image Shadow Removal Self-Supervised Learning +1

Facial Emotion Recognition with Noisy Multi-task Annotations

1 code implementation19 Oct 2020 Siwei Zhang, Zhiwu Huang, Danda Pani Paudel, Luc van Gool

In our formulation, we exploit a new method to enable the emotion prediction and the joint distribution learning in a unified adversarial learning game.

Emotion Recognition

Self-Supervised Ranking for Representation Learning

no code implementations14 Oct 2020 Ali Varamesh, Ali Diba, Tinne Tuytelaars, Luc van Gool

We present a new framework for self-supervised representation learning by formulating it as a ranking problem in an image retrieval context on a large number of random views (augmentations) obtained from images.

Contrastive Learning Image Retrieval +3

SMILE: Semantically-guided Multi-attribute Image and Layout Editing

1 code implementation5 Oct 2020 Andrés Romero, Luc van Gool, Radu Timofte

Additionally, our method is capable of adding, removing or changing either fine-grained or coarse attributes by using an image as a reference or by exploring the style distribution space, and it can be easily extended to head-swapping and face-reenactment applications without being trained on videos.

Face Reenactment Image Manipulation

Few-Shot Classification By Few-Iteration Meta-Learning

1 code implementation1 Oct 2020 Ardhendu Shekhar Tripathi, Martin Danelljan, Luc van Gool, Radu Timofte

By employing an efficient initialization module and a Steepest Descent based optimization algorithm, our base learner predicts a powerful classifier within only a few iterations.

Classification General Classification +2

Depth Estimation from Monocular Images and Sparse Radar Data

1 code implementation30 Sep 2020 Juan-Ting Lin, Dengxin Dai, Luc van Gool

We give a comprehensive study of the fusion between RGB images and Radar measurements from different aspects and proposed a working solution based on the observations.

Depth Estimation

Improving Point Cloud Semantic Segmentation by Learning 3D Object Detection

no code implementations22 Sep 2020 Ozan Unal, Luc van Gool, Dengxin Dai

Point cloud semantic segmentation plays an essential role in autonomous driving, providing vital information about drivable surfaces and nearby objects that can aid higher level tasks such as path planning and collision avoidance.

3D Object Detection 3D Semantic Segmentation +2

Plug-and-Play Image Restoration with Deep Denoiser Prior

4 code implementations31 Aug 2020 Kai Zhang, Yawei Li, WangMeng Zuo, Lei Zhang, Luc van Gool, Radu Timofte

Recent works on plug-and-play image restoration have shown that a denoiser can implicitly serve as the image prior for model-based methods to solve many inverse problems.

Deblurring Demosaicking +1

Learning Condition Invariant Features for Retrieval-Based Localization from 1M Images

1 code implementation27 Aug 2020 Janine Thoma, Danda Pani Paudel, Ajad Chhatkuli, Luc van Gool

Image features for retrieval-based localization must be invariant to dynamic objects (e. g. cars) as well as seasonal and daytime changes.

Automated Search for Resource-Efficient Branched Multi-Task Networks

1 code implementation24 Aug 2020 David Bruggemann, Menelaos Kanakis, Stamatios Georgoulis, Luc van Gool

The multi-modal nature of many vision problems calls for neural network architectures that can perform multiple tasks concurrently.

Neural Architecture Search

Neural Architecture Search as Sparse Supernet

no code implementations31 Jul 2020 Yan Wu, Aoming Liu, Zhiwu Huang, Siwei Zhang, Luc van Gool

This paper aims at enlarging the problem of Neural Architecture Search (NAS) from Single-Path and Multi-Path Search to automated Mixed-Path Search.

Neural Architecture Search

Reparameterizing Convolutions for Incremental Multi-Task Learning without Task Interference

1 code implementation ECCV 2020 Menelaos Kanakis, David Bruggemann, Suman Saha, Stamatios Georgoulis, Anton Obukhov, Luc van Gool

First, enabling the model to be inherently incremental, continuously incorporating information from new tasks without forgetting the previously learned ones (incremental learning).

Incremental Learning Multi-Task Learning

Weakly Supervised 3D Object Detection from Lidar Point Cloud

1 code implementation ECCV 2020 Qinghao Meng, Wenguan Wang, Tianfei Zhou, Jianbing Shen, Luc van Gool, Dengxin Dai

This work proposes a weakly supervised approach for 3D object detection, only requiring a small set of weakly annotated scenes, associated with a few precisely labeled object instances.

3D Object Detection

Video Object Segmentation with Episodic Graph Memory Networks

1 code implementation ECCV 2020 Xiankai Lu, Wenguan Wang, Martin Danelljan, Tianfei Zhou, Jianbing Shen, Luc van Gool

How to make a segmentation model efficiently adapt to a specific video and to online target appearance variations are fundamentally crucial issues in the field of video object segmentation.

Semantic Segmentation Video Object Segmentation +2

Learning Accurate and Human-Like Driving using Semantic Maps and Attention

no code implementations10 Jul 2020 Simon Hecker, Dengxin Dai, Alexander Liniger, Luc van Gool

This paper investigates how end-to-end driving models can be improved to drive more accurately and human-like.

Modelling the Distribution of 3D Brain MRI using a 2D Slice VAE

1 code implementation9 Jul 2020 Anna Volokitin, Ertunc Erdil, Neerav Karani, Kerem Can Tezcan, Xiaoran Chen, Luc van Gool, Ender Konukoglu

We propose a method to model 3D MR brain volumes distribution by combining a 2D slice VAE with a Gaussian model that captures the relationships between slices.

Self-Calibration Supported Robust Projective Structure-from-Motion

no code implementations4 Jul 2020 Rui Gong, Danda Pani Paudel, Ajad Chhatkuli, Luc van Gool

In this paper, we propose a unified SfM method, in which the matching process is supported by self-calibration constraints.

Structure from Motion

The Heterogeneity Hypothesis: Finding Layer-Wise Differentiated Network Architectures

1 code implementation CVPR 2021 Yawei Li, Wen Li, Martin Danelljan, Kai Zhang, Shuhang Gu, Luc van Gool, Radu Timofte

Based on that, we articulate the heterogeneity hypothesis: with the same training protocol, there exists a layer-wise differentiated network architecture (LW-DNA) that can outperform the original network with regular channel configurations but with a lower level of model complexity.

Image Classification Image Restoration +1

OpenDVC: An Open Source Implementation of the DVC Video Compression Method

4 code implementations29 Jun 2020 Ren Yang, Luc van Gool, Radu Timofte

At the time of writing this report, several learned video compression methods are superior to DVC, but currently none of them provides open source codes.


Analogical Image Translation for Fog Generation

no code implementations28 Jun 2020 Rui Gong, Dengxin Dai, Yu-Hua Chen, Wen Li, Luc van Gool

AIT achieves this zero-shot image translation capability by coupling a supervised training scheme in the synthetic domain, a cycle consistency strategy in the real domain, an adversarial training scheme between the two domains, and a novel network design.

Image-to-Image Translation Scene Understanding +1

SRFlow: Learning the Super-Resolution Space with Normalizing Flow

3 code implementations ECCV 2020 Andreas Lugmayr, Martin Danelljan, Luc van Gool, Radu Timofte

SRFlow therefore directly accounts for the ill-posed nature of the problem, and learns to predict diverse photo-realistic high-resolution images.

Image Manipulation Super-Resolution

Learning for Video Compression with Recurrent Auto-Encoder and Recurrent Probability Model

3 code implementations24 Jun 2020 Ren Yang, Fabian Mentzer, Luc van Gool, Radu Timofte

The experiments show that our approach achieves the state-of-the-art learned video compression performance in terms of both PSNR and MS-SSIM.


Consistency Guided Scene Flow Estimation

no code implementations ECCV 2020 Yuhua Chen, Luc van Gool, Cordelia Schmid, Cristian Sminchisescu

To handle inherent modeling error in the consistency loss (e. g. Lambertian assumptions) and for better generalization, we further introduce a learned, output refinement network, which takes the initial predictions, the loss, and the gradient as input, and efficiently predicts a correlated output update.

Scene Flow Estimation

Dense Non-Rigid Structure from Motion: A Manifold Viewpoint

no code implementations15 Jun 2020 Suryansh Kumar, Luc van Gool, Carlos E. P. de Oliveira, Anoop Cherian, Yuchao Dai, Hongdong Li

Assuming that a deforming shape is composed of a union of local linear subspace and, span a global low-rank space over multiple frames enables us to efficiently model complex non-rigid deformations.

Structure from Motion

Map-Guided Curriculum Domain Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation

no code implementations28 May 2020 Christos Sakaridis, Dengxin Dai, Luc van Gool

Our central contributions are: 1) a curriculum framework to gradually adapt semantic segmentation models from day to night through progressively darker times of day, exploiting cross-time-of-day correspondences between daytime images from a reference map and dark images to guide the label inference in the dark domains; 2) a novel uncertainty-aware annotation and evaluation framework and metric for semantic segmentation, including image regions beyond human recognition capability in the evaluation in a principled fashion; 3) the Dark Zurich dataset, comprising 2416 unlabeled nighttime and 2920 unlabeled twilight images with correspondences to their daytime counterparts plus a set of 201 nighttime images with fine pixel-level annotations created with our protocol, which serves as a first benchmark for our novel evaluation.

Domain Adaptation Semantic Segmentation

Safe Motion Planning for Autonomous Driving using an Adversarial Road Model

1 code implementation15 May 2020 Alexander Liniger, Luc van Gool

This formulation allows us to compute safe sets using tools from viability theory, that can be used as terminal constraints in an optimization-based motion planner.

Robotics Systems and Control Systems and Control Optimization and Control

Flexible Example-based Image Enhancement with Task Adaptive Global Feature Self-Guided Network

no code implementations13 May 2020 Dario Kneubuehler, Shuhang Gu, Luc van Gool, Radu Timofte

We propose the first practical multitask image enhancement network, that is able to learn one-to-many and many-to-one image mappings.

Image Enhancement

Action Sequence Predictions of Vehicles in Urban Environments using Map and Social Context

no code implementations29 Apr 2020 Jan-Nico Zaech, Dengxin Dai, Alexander Liniger, Luc van Gool

Our second contribution lies in applying the method to the well-known traffic agent tracking and prediction dataset Argoverse, resulting in 228, 000 action sequences.

Multi-Task Learning for Dense Prediction Tasks: A Survey

1 code implementation28 Apr 2020 Simon Vandenhende, Stamatios Georgoulis, Wouter Van Gansbeke, Marc Proesmans, Dengxin Dai, Luc van Gool

In this survey, we provide a well-rounded view on state-of-the-art deep learning approaches for MTL in computer vision, explicitly emphasizing on dense prediction tasks.

Multi-Task Learning

SESAME: Semantic Editing of Scenes by Adding, Manipulating or Erasing Objects

1 code implementation ECCV 2020 Evangelos Ntavelis, Andrés Romero, Iason Kastanis, Luc van Gool, Radu Timofte

In contrast to previous methods that employ a discriminator that trivially concatenates semantics and image as an input, the SESAME discriminator is composed of two input streams that independently process the image and its semantics, using the latter to manipulate the results of the former.

 Ranked #1 on Image-to-Image Translation on Cityscapes Labels-to-Photo (Per-pixel Accuracy metric)

Image Manipulation Image-to-Image Translation

Quantifying Data Augmentation for LiDAR based 3D Object Detection

no code implementations3 Apr 2020 Martin Hahner, Dengxin Dai, Alexander Liniger, Luc van Gool

In this work, we shed light on different data augmentation techniques commonly used in Light Detection and Ranging (LiDAR) based 3D Object Detection.

3D Object Detection Data Augmentation +1

Learning Unsupervised Hierarchical Part Decomposition of 3D Objects from a Single RGB Image

1 code implementation CVPR 2020 Despoina Paschalidou, Luc van Gool, Andreas Geiger

Humans perceive the 3D world as a set of distinct objects that are characterized by various low-level (geometry, reflectance) and high-level (connectivity, adjacency, symmetry) properties.

3D Reconstruction Hierarchical structure

DHP: Differentiable Meta Pruning via HyperNetworks

2 code implementations ECCV 2020 Yawei Li, Shuhang Gu, Kai Zhang, Luc van Gool, Radu Timofte

Passing the sparsified latent vectors through the hypernetworks, the corresponding slices of the generated weight parameters can be removed, achieving the effect of network pruning.

Denoising Image Classification +3

Probabilistic Regression for Visual Tracking

1 code implementation CVPR 2020 Martin Danelljan, Luc van Gool, Radu Timofte

In this work, we therefore propose a probabilistic regression formulation and apply it to tracking.

Visual Tracking

Learning What to Learn for Video Object Segmentation

2 code implementations ECCV 2020 Goutam Bhat, Felix Järemo Lawin, Martin Danelljan, Andreas Robinson, Michael Felsberg, Luc van Gool, Radu Timofte

This allows us to achieve a rich internal representation of the target in the current frame, significantly increasing the segmentation accuracy of our approach.

Few-Shot Learning One-shot visual object segmentation +3

Know Your Surroundings: Exploiting Scene Information for Object Tracking

1 code implementation ECCV 2020 Goutam Bhat, Martin Danelljan, Luc van Gool, Radu Timofte

Such approaches are however prone to fail in case of e. g. fast appearance changes or presence of distractor objects, where a target appearance model alone is insufficient for robust tracking.

Object Tracking

Learning Better Lossless Compression Using Lossy Compression

1 code implementation CVPR 2020 Fabian Mentzer, Luc van Gool, Michael Tschannen

We leverage the powerful lossy image compression algorithm BPG to build a lossless image compression system.

Image Compression

Deep Unfolding Network for Image Super-Resolution

1 code implementation CVPR 2020 Kai Zhang, Luc van Gool, Radu Timofte

As a result, the proposed network inherits the flexibility of model-based methods to super-resolve blurry, noisy images for different scale factors via a single model, while maintaining the advantages of learning-based methods.

Image Super-Resolution

Geometrically Mappable Image Features

1 code implementation21 Mar 2020 Janine Thoma, Danda Pani Paudel, Ajad Chhatkuli, Luc van Gool

This is achieved by guiding the learning process such that the feature and geometric distances between images are directly proportional.

Image Retrieval

Group Sparsity: The Hinge Between Filter Pruning and Decomposition for Network Compression

2 code implementations CVPR 2020 Yawei Li, Shuhang Gu, Christoph Mayer, Luc van Gool, Radu Timofte

In this paper, we analyze two popular network compression techniques, i. e. filter pruning and low-rank decomposition, in a unified sense.

Unsupervised Learning of Category-Specific Symmetric 3D Keypoints from Point Sets

1 code implementation ECCV 2020 Clara Fernandez-Labrador, Ajad Chhatkuli, Danda Pani Paudel, Jose J. Guerrero, Cédric Demonceaux, Luc van Gool

This paper aims at learning category-specific 3D keypoints, in an unsupervised manner, using a collection of misaligned 3D point clouds of objects from an unknown category.

Semantic Object Prediction and Spatial Sound Super-Resolution with Binaural Sounds

no code implementations ECCV 2020 Arun Balajee Vasudevan, Dengxin Dai, Luc van Gool

We also propose two auxiliary tasks namely, a) a novel task on Spatial Sound Super-resolution to increase the spatial resolution of sounds, and b) dense depth prediction of the scene.

Depth Estimation Super-Resolution

Learning for Video Compression with Hierarchical Quality and Recurrent Enhancement

3 code implementations CVPR 2020 Ren Yang, Fabian Mentzer, Luc van Gool, Radu Timofte

In our HLVC approach, the hierarchical quality benefits the coding efficiency, since the high quality information facilitates the compression and enhancement of low quality frames at encoder and decoder sides, respectively.

Image Compression MS-SSIM +2

Replacing Mobile Camera ISP with a Single Deep Learning Model

2 code implementations13 Feb 2020 Andrey Ignatov, Luc van Gool, Radu Timofte

The model is trained to convert RAW Bayer data obtained directly from mobile camera sensor into photos captured with a professional high-end DSLR camera, making the solution independent of any particular mobile ISP implementation.

Demosaicking Denoising

MTI-Net: Multi-Scale Task Interaction Networks for Multi-Task Learning

1 code implementation ECCV 2020 Simon Vandenhende, Stamatios Georgoulis, Luc van Gool

In this paper, we argue about the importance of considering task interactions at multiple scales when distilling task information in a multi-task learning setup.

Multi-Task Learning Semantic Segmentation

Don't Forget The Past: Recurrent Depth Estimation from Monocular Video

no code implementations8 Jan 2020 Vaishakh Patil, Wouter Van Gansbeke, Dengxin Dai, Luc van Gool

In particular, we put three different types of depth estimation (supervised depth prediction, self-supervised depth prediction, and self-supervised depth completion) into a common framework.

Depth Completion Monocular Depth Estimation +1

Efficient Video Semantic Segmentation with Labels Propagation and Refinement

no code implementations26 Dec 2019 Matthieu Paul, Christoph Mayer, Luc van Gool, Radu Timofte

(ii) On the GPU, two Convolutional Neural Networks: A main segmentation network that is used to predict dense semantic labels from scratch, and a Refiner that is designed to improve predictions from previous frames with the help of a fast Inconsistencies Attention Module (IAM).

Optical Flow Estimation Real-Time Semantic Segmentation +2

Domain Agnostic Feature Learning for Image and Video Based Face Anti-spoofing

no code implementations15 Dec 2019 Suman Saha, Wen-Hao Xu, Menelaos Kanakis, Stamatios Georgoulis, Yu-Hua Chen, Danda Pani Paudel, Luc van Gool

Face anti-spoofing is a measure towards this direction for bio-metric user authentication, and in particular face recognition, that tries to prevent spoof attacks.

Face Anti-Spoofing Face Recognition

Towards Partial Supervision for Generic Object Counting in Natural Scenes

1 code implementation13 Dec 2019 Hisham Cholakkal, Guolei Sun, Salman Khan, Fahad Shahbaz Khan, Ling Shao, Luc van Gool

Our RLC framework further reduces the annotation cost arising from large numbers of object categories in a dataset by only using lower-count supervision for a subset of categories and class-labels for the remaining ones.

Image Classification Instance Segmentation +2

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 +4

Divide-and-Conquer Adversarial Learning for High-Resolution Image and Video Enhancement

no code implementations23 Oct 2019 Zhiwu Huang, Danda Pani Paudel, Guanju Li, Jiqing Wu, Radu Timofte, Luc van Gool

This paper introduces a divide-and-conquer inspired adversarial learning (DACAL) approach for photo enhancement.

Video Enhancement

AI Benchmark: All About Deep Learning on Smartphones in 2019

no code implementations15 Oct 2019 Andrey Ignatov, Radu Timofte, Andrei Kulik, Seungsoo Yang, Ke Wang, Felix Baum, Max Wu, Lirong Xu, Luc van Gool

The performance of mobile AI accelerators has been evolving rapidly in the past two years, nearly doubling with each new generation of SoCs.

Convex Relaxations for Consensus and Non-Minimal Problems in 3D Vision

no code implementations ICCV 2019 Thomas Probst, Danda Pani Paudel, Ajad Chhatkuli, Luc van Gool

Notably, we further exploit the POP formulation of non-minimal solver also for the generic consensus maximization problems in 3D vision.

Structure from Motion

Talk2Car: Taking Control of Your Self-Driving Car

1 code implementation IJCNLP 2019 Thierry Deruyttere, Simon Vandenhende, Dusan Grujicic, Luc van Gool, Marie-Francine Moens

Or more specifically, we consider the problem in an autonomous driving setting, where a passenger requests an action that can be associated with an object found in a street scene.

Autonomous Driving Referring Expression Comprehension +1

Extremely Weak Supervised Image-to-Image Translation for Semantic Segmentation

1 code implementation18 Sep 2019 Samarth Shukla, Luc van Gool, Radu Timofte

Recent advances in generative models and adversarial training have led to a flourishing image-to-image (I2I) translation literature.

Image-to-Image Translation Semantic Segmentation +1

Texture Underfitting for Domain Adaptation

no code implementations29 Aug 2019 Jan-Nico Zaech, Dengxin Dai, Martin Hahner, Luc van Gool

Comprehensive semantic segmentation is one of the key components for robust scene understanding and a requirement to enable autonomous driving.

Autonomous Driving Domain Adaptation +3

Learning Filter Basis for Convolutional Neural Network Compression

3 code implementations ICCV 2019 Yawei Li, Shuhang Gu, Luc van Gool, Radu Timofte

Convolutional neural networks (CNNs) based solutions have achieved state-of-the-art performances for many computer vision tasks, including classification and super-resolution of images.

General Classification Image Classification +2

Dual Grid Net: hand mesh vertex regression from single depth maps

no code implementations ECCV 2020 Chengde Wan, Thomas Probst, Luc van Gool, Angela Yao

In the first stage, the network estimates a dense correspondence field for every pixel on the depth map or image grid to the mesh grid.

Learning a Curve Guardian for Motorcycles

no code implementations12 Jul 2019 Simon Hecker, Alexander Liniger, Henrik Maurenbrecher, Dengxin Dai, Luc van Gool

Our contributes are fourfold: 1) we predict the motorcycle's intra-lane position using a convolutional neural network (CNN), 2) we predict the motorcycle roll angle using a CNN, 3) we use an upgraded controller model that incorporates road incline for a more realistic model and prediction, 4) we design a scale-able system by utilizing HERE Technologies map database to obtain the accurate road geometry of the future path.

Gated CRF Loss for Weakly Supervised Semantic Image Segmentation

no code implementations11 Jun 2019 Anton Obukhov, Stamatios Georgoulis, Dengxin Dai, Luc van Gool

State-of-the-art approaches for semantic segmentation rely on deep convolutional neural networks trained on fully annotated datasets, that have been shown to be notoriously expensive to collect, both in terms of time and money.

Weakly-Supervised Semantic Segmentation

3D Appearance Super-Resolution with Deep Learning

1 code implementation CVPR 2019 Yawei Li, Vagia Tsiminaki, Radu Timofte, Marc Pollefeys, Luc van Gool

Experimental results demonstrate that our proposed networks successfully incorporate the 3D geometric information and super-resolve the texture maps.


Semi-Supervised Learning by Augmented Distribution Alignment

1 code implementation ICCV 2019 Qin Wang, Wen Li, Luc van Gool

We reveal that an essential sampling bias exists in semi-supervised learning due to the limited number of labeled samples, which often leads to a considerable empirical distribution mismatch between labeled data and unlabeled data.

Domain Adaptation Semi-Supervised Image Classification

The 2019 DAVIS Challenge on VOS: Unsupervised Multi-Object Segmentation

no code implementations2 May 2019 Sergi Caelles, Jordi Pont-Tuset, Federico Perazzi, Alberto Montes, Kevis-Kokitsi Maninis, Luc van Gool

We present the 2019 DAVIS Challenge on Video Object Segmentation, the third edition of the DAVIS Challenge series, a public competition designed for the task of Video Object Segmentation (VOS).

Semantic Segmentation Video Object Segmentation +1

DynamoNet: Dynamic Action and Motion Network

no code implementations ICCV 2019 Ali Diba, Vivek Sharma, Luc van Gool, Rainer Stiefelhagen

With these overall objectives, to this end, we introduce a novel unified spatio-temporal 3D-CNN architecture (DynamoNet) that jointly optimizes the video classification and learning motion representation by predicting future frames as a multi-task learning problem.

Action Recognition Classification +3

Large Scale Holistic Video Understanding

1 code implementation ECCV 2020 Ali Diba, Mohsen Fayyaz, Vivek Sharma, Manohar Paluri, Jurgen Gall, Rainer Stiefelhagen, Luc van Gool

HVU is organized hierarchically in a semantic taxonomy that focuses on multi-label and multi-task video understanding as a comprehensive problem that encompasses the recognition of multiple semantic aspects in the dynamic scene.

Action Classification Action Recognition +5

A Novel BiLevel Paradigm for Image-to-Image Translation

no code implementations18 Apr 2019 Liqian Ma, Qianru Sun, Bernt Schiele, Luc van Gool

Image-to-image (I2I) translation is a pixel-level mapping that requires a large number of paired training data and often suffers from the problems of high diversity and strong category bias in image scenes.

Image-to-Image Translation Translation

Learning Discriminative Model Prediction for Tracking

2 code implementations ICCV 2019 Goutam Bhat, Martin Danelljan, Luc van Gool, Radu Timofte

The current strive towards end-to-end trainable computer vision systems imposes major challenges for the task of visual tracking.

Visual Object Tracking Visual Tracking

Sliced Wasserstein Generative Models

1 code implementation CVPR 2019 Jiqing Wu, Zhiwu Huang, Dinesh Acharya, Wen Li, Janine Thoma, Danda Pani Paudel, Luc van Gool

In generative modeling, the Wasserstein distance (WD) has emerged as a useful metric to measure the discrepancy between generated and real data distributions.

Image Generation Video Generation

Branched Multi-Task Networks: Deciding What Layers To Share

no code implementations ICLR 2020 Simon Vandenhende, Stamatios Georgoulis, Bert de Brabandere, Luc van Gool

In the context of multi-task learning, neural networks with branched architectures have often been employed to jointly tackle the tasks at hand.

Multi-Task Learning Neural Architecture Search

Fast video object segmentation with Spatio-Temporal GANs

no code implementations28 Mar 2019 Sergi Caelles, Albert Pumarola, Francesc Moreno-Noguer, Alberto Sanfeliu, Luc van Gool

To achieve this, we concentrate all the heavy computational load to the training phase with two critics that enforce spatial and temporal mask consistency over the last K frames.

Fine-tuning One-shot visual object segmentation +3

Learning Accurate, Comfortable and Human-like Driving

no code implementations26 Mar 2019 Simon Hecker, Dengxin Dai, Luc van Gool

Our model is trained and evaluated on the Drive360 dataset, which features 60 hours and 3000 km of real-world driving data.

Autonomous Vehicles

A Three-Player GAN: Generating Hard Samples To Improve Classification Networks

no code implementations8 Mar 2019 Simon Vandenhende, Bert de Brabandere, Davy Neven, Luc van Gool

The generator's objective is to synthesize samples that are both realistic and hard to label for the classifier.