no code implementations • ECCV 2020 • Wei Zeng, Sezer Karaoglu, Theo Gevers
Leveraging the layout depth map as an intermediate representation, our proposed method outperforms existing methods for both panorama layout prediction and depth estimation.
no code implementations • 29 Mar 2024 • Qi Bi, ShaoDi You, Theo Gevers
In this paper, we start with solid revisit of the physics definition of weather and how it can be described as a continuous machine learning and computer vision task.
1 code implementation • Association for the Advancement of Artificial Intelligence (AAAI) 2024 • Qi Bi, ShaoDi You, Theo Gevers
We argue that an ideal segmentation model that can be well generalized to foggy-scenes need to simultaneously enhance the content, de-correlate the urban-scene style and de-correlate the fog style.
no code implementations • 15 Dec 2023 • Weijie Wei, Fatemeh Karimi Nejadasl, Theo Gevers, Martin R. Oswald
The scarcity of annotated data in LiDAR point cloud understanding hinders effective representation learning.
no code implementations • 6 Dec 2023 • Vladimir Yugay, Yue Li, Theo Gevers, Martin R. Oswald
We present a dense simultaneous localization and mapping (SLAM) method that uses 3D Gaussians as a scene representation.
no code implementations • 11 Oct 2023 • Rick Groenendijk, Leo Dorst, Theo Gevers
In the network, morphological Haar sampling is applied to both feature channels in several layers, which splits extreme values and high-frequency information such that both can be processed to improve both modalities.
Ranked #47 on Semantic Segmentation on NYU Depth v2
1 code implementation • 11 Oct 2023 • Osman Ülger, Yu Wang, Ysbrand Galama, Sezer Karaoglu, Theo Gevers, Martin R. Oswald
Humans have a remarkable ability to perceive and reason about the world around them by understanding the relationships between objects.
1 code implementation • 29 Sep 2023 • Weijie Wei, Martin R. Oswald, Fatemeh Karimi Nejadasl, Theo Gevers
To leverage the different properties of each branch, we employ a geometry-aware fusion module that is learned to combine the results of each branch.
1 code implementation • 20 Jul 2023 • Xiaoyan Xing, Konrad Groh, Sezer Karaoglu, Theo Gevers
The purpose of intrinsic decomposition is to separate an image into its albedo (reflective properties) and shading components (illumination properties).
1 code implementation • IEEE Transactions on Image Processing 2023 • Qi Bi, ShaoDi You, Theo Gevers
In this paper, in contrast to existing methods, we tackle this challenge from the perspective of image formulation itself, where the image appearance is determined by both intrinsic (e. g., semantic category, structure) and extrinsic (e. g., lighting) properties.
Ranked #1 on Semantic Segmentation on Mapillary val
1 code implementation • 1 Jul 2023 • Qi Bi, ShaoDi You, Theo Gevers
Unlike domain gap challenges, USSS is unique in that the semantic categories are often similar in different urban scenes, while the styles can vary significantly due to changes in urban landscapes, weather conditions, lighting, and other factors.
no code implementations • ICCV 2023 • Ruihong Yin, Sezer Karaoglu, Theo Gevers
First, geometry-guided feature learning encodes geometric priors to contain view-dependent information.
1 code implementation • 25 Nov 2022 • Rick Groenendijk, Leo Dorst, Theo Gevers
Pooling is essentially an operation from the field of Mathematical Morphology, with max pooling as a limited special case.
1 code implementation • 30 Aug 2022 • Partha Das, Sezer Karaoglu, Arjan Gijsenij, Theo Gevers
An ablation study is conducted showing that the use of the proposed priors and progressive CNN increase the IID performance.
no code implementations • Computer Vision and Image Understanding 2022 • YaHui Zhang, ShaoDi You, Sezer Karaoglu, Theo Gevers
Multi-person 3D pose estimation with absolute depths for a fisheye camera is a challenging task but with valuable applications in daily life, especially for video surveillance.
no code implementations • 8 Apr 2022 • Anil S. Baslamisli, Theo Gevers
We improve upon their model by introducing illumination invariant image descriptors: color ratios.
1 code implementation • CVPR 2022 • Partha Das, Sezer Karaoglu, Theo Gevers
An extensive ablation study and large scale experiments are conducted showing that it is beneficial for edge-driven hybrid IID networks to make use of illumination invariant descriptors and that separating global and local cues helps in improving the performance of the network.
no code implementations • 2 Sep 2021 • Partha Das, Yang Liu, Sezer Karaoglu, Theo Gevers
However, most of the existing color constancy methods are designed for single light sources.
1 code implementation • 9 Nov 2020 • Hoang-An Le, Thomas Mensink, Partha Das, Sezer Karaoglu, Theo Gevers
Multimodal large-scale datasets for outdoor scenes are mostly designed for urban driving problems.
no code implementations • 22 Oct 2020 • Ipek Ganiyusufoglu, L. Minh Ngô, Nedko Savov, Sezer Karaoglu, Theo Gevers
In this paper, we empirically show that existing approaches on image and sequence classifiers generalize poorly to new manipulation techniques.
1 code implementation • 17 Sep 2020 • Hoang-An Le, Thomas Mensink, Partha Das, Theo Gevers
In this paper the argument is made that for true novel view synthesis of objects, where the object can be synthesized from any viewpoint, an explicit 3D shape representation isdesired.
no code implementations • 3 Sep 2020 • Anil S. Baslamisli, Yang Liu, Sezer Karaoglu, Theo Gevers
We investigate the use of photometric invariance and deep learning to compute intrinsic images (albedo and shading).
1 code implementation • 3 Sep 2020 • Rick Groenendijk, Sezer Karaoglu, Theo Gevers, Thomas Mensink
In this paper, we propose a weighting scheme based on the coefficient of variations and set the weights based on properties observed while training the model.
1 code implementation • ECCV 2020 • Wei Wang, ShaoDi You, Sezer Karaoglu, Theo Gevers
The experiments further show significant performance improvement of kinship verification when trained on the same dataset with more realistic distributions.
no code implementations • 9 Dec 2019 • Anil S. Baslamisli, Partha Das, Hoang-An Le, Sezer Karaoglu, Theo Gevers
The aim is to distinguish strong photometric effects from reflectance variations.
1 code implementation • 29 Oct 2019 • Rick Groenendijk, Sezer Karaoglu, Theo Gevers, Thomas Mensink
For the quality of the image reconstruction and disparity prediction, a combination of different losses is used, including L1 image reconstruction losses and left-right disparity smoothness.
no code implementations • 26 Dec 2018 • Minh Ngô, Burak Mandira, Selim Fırat Yılmaz, Ward Heij, Sezer Karaoglu, Henri Bouma, Hamdi Dibeklioglu, Theo Gevers
Lies and deception are common phenomena in society, both in our private and professional lives.
no code implementations • 18 Dec 2018 • Jian Han, Sezer Karaoglu, Hoang-An Le, Theo Gevers
In this paper, we provide a synthetic data generator methodology with fully controlled, multifaceted variations based on a new 3D face dataset (3DU-Face).
no code implementations • 7 Dec 2018 • Partha Das, Anil S. Baslamisli, Yang Liu, Sezer Karaoglu, Theo Gevers
In this paper, we formulate the color constancy task as an image-to-image translation problem using GANs.
1 code implementation • 5 Dec 2018 • Hoàng-Ân Lê, Tushar Nimbhorkar, Thomas Mensink, Anil S. Baslamisli, Sezer Karaoglu, Theo Gevers
There hardly exists any large-scale datasets with dense optical flow of non-rigid motion from real-world imagery as of today.
no code implementations • 4 Dec 2018 • Wei Zeng, Sezer Karaoglu, Theo Gevers
In this paper, we propose a pipeline to generate 3D point cloud of an object from a single-view RGB image.
1 code implementation • ECCV 2018 • Anil S. Baslamisli, Thomas T. Groenestege, Partha Das, Hoang-An Le, Sezer Karaoglu, Theo Gevers
To that end, we propose a supervised end-to-end CNN architecture to jointly learn intrinsic image decomposition and semantic segmentation.
1 code implementation • 19 Jul 2018 • Hoang-An Le, Anil S. Baslamisli, Thomas Mensink, Theo Gevers
Optical flow, semantic segmentation, and surface normals represent different information modalities, yet together they bring better cues for scene understanding problems.
no code implementations • CVPR 2018 • Anil S. Baslamisli, Hoang-An Le, Theo Gevers
On the other hand, recent research use deep learning models as in-and-out black box and do not consider the well-established, traditional image formation process as the basis of their intrinsic learning process.
no code implementations • 30 Nov 2017 • Wei Zeng, Theo Gevers
Classification and segmentation of 3D point clouds are important tasks in computer vision.
no code implementations • 26 Dec 2014 • Sezer Karaoglu, Yang Liu, Theo Gevers
Experiments on the PASCAL VOC07 and VOC10 datasets show that the proposed method significantly outperforms single object detectors, DPM (8. 4%), CN (6. 8%) and EES (17. 0%) on VOC07 and DPM (6. 5%), CN (5. 5%) and EES (16. 2%) on VOC10.
no code implementations • 11 Dec 2014 • Jose M. Alvarez, Theo Gevers, Antonio M. Lopez
These algorithms reduce the effect of lighting variations and weather conditions by exploiting the discriminant/invariant properties of different color representations.
no code implementations • CVPR 2013 • Ivo Everts, Jan C. van Gemert, Theo Gevers
Existing STIP-based action recognition approaches operate on intensity representations of the image data.