1 code implementation • 14 Apr 2025 • Raphael Achddou, Yann Gousseau, Saïd Ladjal, Sabine Süsstrunk
One way to address these shortcomings is to have a better control over the training sets, in particular by using synthetic sets.
no code implementations • 29 Mar 2025 • Yufan Ren, Konstantinos Tertikas, Shalini Maiti, Junlin Han, Tong Zhang, Sabine Süsstrunk, Filippos Kokkinos
Our results reveal that even the state-of-the-art LVLMs struggle with these puzzles, highlighting fundamental limitations in their puzzle-solving capabilities.
no code implementations • 14 Mar 2025 • Liying Lu, Raphaël Achddou, Sabine Süsstrunk
Low-light photography produces images with low signal-to-noise ratios due to limited photons.
no code implementations • 5 Feb 2025 • Dongqing Wang, Ehsan Pajouheshgar, Yitao Xu, Tong Zhang, Sabine Süsstrunk
Artistic stylization of 3D volumetric smoke data is still a challenge in computer graphics due to the difficulty of ensuring spatiotemporal consistency given a reference style image, and that within reasonable time and computational resources.
1 code implementation • 27 Oct 2024 • Peter Grönquist, Deblina Bhattacharjee, Bahar Aydemir, Baran Ozaydin, Tong Zhang, Mathieu Salzmann, Sabine Süsstrunk
This dataset is a crucial component of the AI4VA Workshop Challenges~\url{https://sites. google. com/view/ai4vaeccv2024}, where we specifically explore depth and saliency.
1 code implementation • 11 Sep 2024 • Bahar Aydemir, Deblina Bhattacharjee, Tong Zhang, Mathieu Salzmann, Sabine Süsstrunk
We propose a novel data augmentation method for deep saliency prediction that edits natural images while preserving the complexity and variability of real-world scenes.
no code implementations • 11 Jul 2024 • Shuangqi Li, Chen Liu, Tong Zhang, Hieu Le, Sabine Süsstrunk, Mathieu Salzmann
We introduce an approach to bias deep generative models, such as GANs and diffusion models, towards generating data with either enhanced fidelity or increased diversity.
1 code implementation • 10 Jul 2024 • Lingzhi Pan, Tong Zhang, Bingyuan Chen, Qi Zhou, Wei Ke, Sabine Süsstrunk, Mathieu Salzmann
Our method searches latent spaces capable of generating inpainted regions that exhibit high fidelity to user-provided prompts while maintaining coherence with the background.
no code implementations • 12 Jun 2024 • Yitao Xu, Tong Zhang, Sabine Süsstrunk
In this paper, we propose Adaptor Neural Cellular Automata (AdaNCA) for Vision Transformers that uses NCA as plug-and-play adaptors between ViT layers, thus enhancing ViT's performance and robustness against adversarial samples as well as out-of-distribution inputs.
1 code implementation • 9 Apr 2024 • Yitao Xu, Ehsan Pajouheshgar, Sabine Süsstrunk
Neural Cellular Automata (NCA) models are trainable variations of traditional Cellular Automata (CA).
no code implementations • 9 Apr 2024 • Ehsan Pajouheshgar, Yitao Xu, Sabine Süsstrunk
We demonstrate the effectiveness of our approach in preserving the consistency of NCA dynamics across a wide range of spatio-temporal granularities.
no code implementations • 11 Mar 2024 • Baran Ozaydin, Tong Zhang, Deblina Bhattacharjee, Sabine Süsstrunk, Mathieu Salzmann
Our OMH yields better unsupervised segmentation performance compared to existing USS methods.
1 code implementation • 5 Dec 2023 • Zhi Chen, Yufan Ren, Tong Zhang, Zheng Dang, Wenbing Tao, Sabine Süsstrunk, Mathieu Salzmann
To achieve this, we introduce a training procedure and a refinement network.
no code implementations • 17 Nov 2023 • Peter Grönquist, Yufan Ren, Qingyi He, Alessio Verardo, Sabine Süsstrunk
Video DeepFakes are fake media created with Deep Learning (DL) that manipulate a person's expression or identity.
no code implementations • 6 Nov 2023 • Ehsan Pajouheshgar, Yitao Xu, Alexander Mordvintsev, Eyvind Niklasson, Tong Zhang, Sabine Süsstrunk
We propose Mesh Neural Cellular Automata (MeshNCA), a method that directly synthesizes dynamic textures on 3D meshes without requiring any UV maps.
no code implementations • 27 Sep 2023 • Martin Nicolas Everaert, Athanasios Fitsios, Marco Bocchio, Sami Arpa, Sabine Süsstrunk, Radhakrishna Achanta
This enables us to generate images with more varied brightness, and images that better match a desired style or color.
no code implementations • ICCV 2023 • Deblina Bhattacharjee, Sabine Süsstrunk, Mathieu Salzmann
We introduce the first multitasking vision transformer adapters that learn generalizable task affinities which can be applied to novel tasks and domains.
no code implementations • 16 Jul 2023 • Deblina Bhattacharjee, Sabine Süsstrunk, Mathieu Salzmann
In this paper, we develop a MultiTask Learning (MTL) model to achieve dense predictions for comics panels to, in turn, facilitate the transfer of comics from one publication channel to another by assisting authors in the task of reconfiguring their narratives.
no code implementations • CVPR 2024 • Dongqing Wang, Tong Zhang, Alaa Abboud, Sabine Süsstrunk
We propose InNeRF360, an automatic system that accurately removes text-specified objects from 360-degree Neural Radiance Fields (NeRF).
1 code implementation • 29 Mar 2023 • Congpei Qiu, Tong Zhang, Wei Ke, Mathieu Salzmann, Sabine Süsstrunk
Dense Self-Supervised Learning (SSL) methods address the limitations of using image-level feature representations when handling images with multiple objects.
1 code implementation • CVPR 2023 • Yanhao Wu, Tong Zhang, Wei Ke, Sabine Süsstrunk, Mathieu Salzmann
In this paper, we introduce an SSL strategy that leverages positive pairs in both the spatial and temporal domain.
no code implementations • ICCV 2023 • Dongqing Wang, Tong Zhang, Sabine Süsstrunk
We propose NEMTO, the first end-to-end neural rendering pipeline to model 3D transparent objects with complex geometry and unknown indices of refraction.
1 code implementation • 5 Jan 2023 • Bahar Aydemir, Ludo Hoffstetter, Tong Zhang, Mathieu Salzmann, Sabine Süsstrunk
Deep saliency prediction algorithms complement the object recognition features, they typically rely on additional information, such as scene context, semantic relationships, gaze direction, and object dissimilarity.
no code implementations • ICCV 2023 • Martin Nicolas Everaert, Marco Bocchio, Sami Arpa, Sabine Süsstrunk, Radhakrishna Achanta
Not adapting this initial latent tensor to the style makes fine-tuning slow, expensive, and impractical, especially when only a few target style images are available.
1 code implementation • CVPR 2023 • Bahar Aydemir, Ludo Hoffstetter, Tong Zhang, Mathieu Salzmann, Sabine Süsstrunk
Deep saliency prediction algorithms complement the object recognition features, they typically rely on additional information such as scene context, semantic relationships, gaze direction, and object dissimilarity.
Ranked #3 on
Saliency Prediction
on SALICON
1 code implementation • 26 Dec 2022 • Baran Ozaydin, Tong Zhang, Sabine Süsstrunk, Mathieu Salzmann
To stylize the source content with the exemplar style, we extract unsupervised cross-domain semantic correspondences and warp the exemplar style to the source content.
1 code implementation • CVPR 2023 • Yufan Ren, Fangjinhua Wang, Tong Zhang, Marc Pollefeys, Sabine Süsstrunk
The success of the Neural Radiance Fields (NeRF) in novel view synthesis has inspired researchers to propose neural implicit scene reconstruction.
no code implementations • CVPR 2023 • Ehsan Pajouheshgar, Yitao Xu, Tong Zhang, Sabine Süsstrunk
Current Dynamic Texture Synthesis (DyTS) models can synthesize realistic videos.
1 code implementation • 10 Oct 2022 • Nicolas Bähler, Majed El Helou, Étienne Objois, Kaan Okumuş, Sabine Süsstrunk
To fill this gap, we derive a novel, cumulant-based, approach for Poisson-Gaussian noise modeling from paired image samples.
1 code implementation • 25 Aug 2022 • Xiaoyu Lin, Baran Ozaydin, Vidit Vidit, Majed El Helou, Sabine Süsstrunk
It would enable drones to fly higher covering larger fields of view, while maintaining a high image quality.
no code implementations • 6 Jun 2022 • Zhichao Huang, Yanbo Fan, Chen Liu, Weizhong Zhang, Yong Zhang, Mathieu Salzmann, Sabine Süsstrunk, Jue Wang
While adversarial training and its variants have shown to be the most effective algorithms to defend against adversarial attacks, their extremely slow training process makes it hard to scale to large datasets like ImageNet.
1 code implementation • CVPR 2022 • Deblina Bhattacharjee, Tong Zhang, Sabine Süsstrunk, Mathieu Salzmann
At the heart of our approach is a shared attention mechanism modeling the dependencies across the tasks.
1 code implementation • CVPR 2022 • Tong Zhang, Congpei Qiu, Wei Ke, Sabine Süsstrunk, Mathieu Salzmann
In essence, this strategy ignores the fact that two crops may truly contain different image information, e. g., background and small objects, and thus tends to restrain the diversity of the learned representations.
1 code implementation • 8 Mar 2022 • Di Chang, Aljaž Božič, Tong Zhang, Qingsong Yan, Yingcong Chen, Sabine Süsstrunk, Matthias Nießner
Finding accurate correspondences among different views is the Achilles' heel of unsupervised Multi-View Stereo (MVS).
1 code implementation • 3 Feb 2022 • Chen Liu, Ziqi Zhao, Sabine Süsstrunk, Mathieu Salzmann
In this paper, we introduce an approach to obtain robust yet compact models by pruning randomly-initialized binary networks.
1 code implementation • 2 Jan 2022 • Qiyuan Liang, Florian Cassayre, Haley Owsianko, Majed El Helou, Sabine Süsstrunk
In this framework, we exploit the outputs of a deep denoising network alongside an image convolved with a reliable filter.
1 code implementation • 14 Dec 2021 • Chen Liu, Zhichao Huang, Mathieu Salzmann, Tong Zhang, Sabine Süsstrunk
This lets us demonstrate that the decay in generalization performance of adversarial training is a result of fitting hard adversarial instances.
1 code implementation • 24 Nov 2021 • Ehsan Pajouheshgar, Tong Zhang, Sabine Süsstrunk
Generative Adversarial Network (GAN) based localized image editing can suffer from ambiguity between semantic attributes.
1 code implementation • 7 Oct 2021 • Deblina Bhattacharjee, Martin Everaert, Mathieu Salzmann, Sabine Süsstrunk
Estimating the depth of comics images is challenging as such images a) are monocular; b) lack ground-truth depth annotations; c) differ across different artistic styles; d) are sparse and noisy.
Ranked #1 on
Depth Estimation
on eBDtheque
no code implementations • 29 Sep 2021 • Zhichao Huang, Chen Liu, Mathieu Salzmann, Sabine Süsstrunk, Tong Zhang
Although adversarial training and its variants currently constitute the most effective way to achieve robustness against adversarial attacks, their poor generalization limits their performance on the test samples.
1 code implementation • 1 Jun 2021 • Xiaoyu Lin, Deblina Bhattacharjee, Majed El Helou, Sabine Süsstrunk
Furthermore, as proof of concept, we show that when using our oracle fidelity map we even outperform the fully retrained methods, whether trained on noisy or restored images.
1 code implementation • 8 Apr 2021 • Bahar Aydemir, Deblina Bhattacharjee, Tong Zhang, Seungryong Kim, Mathieu Salzmann, Sabine Süsstrunk
Saliency prediction has made great strides over the past two decades, with current techniques modeling low-level information, such as color, intensity and size contrasts, and high-level ones, such as attention and gaze direction for entire objects.
1 code implementation • 12 Jan 2021 • Xiaoqi Ma, Xiaoyu Lin, Majed El Helou, Sabine Süsstrunk
While novel denoising networks were designed for real images coming from different distributions, or for specific applications, comparatively small improvement was achieved on Gaussian denoising.
2 code implementations • 3 Nov 2020 • Majed El Helou, Sabine Süsstrunk
Our method, though partly reliant on the quality of the generative network inversion, is competitive with state-of-the-art supervised and task-specific restoration methods.
2 code implementations • 27 Sep 2020 • Majed El Helou, Ruofan Zhou, Sabine Süsstrunk, Radu Timofte, Mahmoud Afifi, Michael S. Brown, Kele Xu, Hengxing Cai, Yuzhong Liu, Li-Wen Wang, Zhi-Song Liu, Chu-Tak Li, Sourya Dipta Das, Nisarg A. Shah, Akashdeep Jassal, Tongtong Zhao, Shanshan Zhao, Sabari Nathan, M. Parisa Beham, R. Suganya, Qing Wang, Zhongyun Hu, Xin Huang, Yaning Li, Maitreya Suin, Kuldeep Purohit, A. N. Rajagopalan, Densen Puthussery, Hrishikesh P. S, Melvin Kuriakose, Jiji C. V, Yu Zhu, Liping Dong, Zhuolong Jiang, Chenghua Li, Cong Leng, Jian Cheng
The first track considered one-to-one relighting; the objective was to relight an input photo of a scene with a different color temperature and illuminant orientation (i. e., light source position).
no code implementations • ECCV 2020 • Seungryong Kim, Sabine Süsstrunk, Mathieu Salzmann
We design our VTN as an encoder-decoder network, with modules dedicated to letting the information flow across the feature channels, to account for the dependencies between the semantic parts.
1 code implementation • NeurIPS 2020 • Chen Liu, Mathieu Salzmann, Tao Lin, Ryota Tomioka, Sabine Süsstrunk
We analyze the influence of adversarial training on the loss landscape of machine learning models.
2 code implementations • 11 May 2020 • Majed El Helou, Ruofan Zhou, Johan Barthas, Sabine Süsstrunk
Deep image relighting is gaining more interest lately, as it allows photo enhancement through illumination-specific retouching without human effort.
2 code implementations • CVPR 2020 • Edo Collins, Raja Bala, Bob Price, Sabine Süsstrunk
Focusing on StyleGAN, we introduce a simple and effective method for making local, semantically-aware edits to a target output image.
no code implementations • 14 Apr 2020 • Majed El Helou, Ruofan Zhou, Frank Schmutz, Fabrice Guibert, Sabine Süsstrunk
Extreme image or video completion, where, for instance, we only retain 1% of pixels in random locations, allows for very cheap sampling in terms of the required pre-processing.
1 code implementation • 19 Mar 2020 • Gökhan Yildirim, Debashis Sen, Mohan Kankanhalli, Sabine Süsstrunk
In this paper, we corroborate based on three subjective experiments on a novel image dataset that objects in natural images are inherently perceived to have varying levels of importance.
2 code implementations • ECCV 2020 • Majed El Helou, Ruofan Zhou, Sabine Süsstrunk
Super-resolution and denoising are ill-posed yet fundamental image restoration tasks.
2 code implementations • 12 Mar 2020 • Ruofan Zhou, Majed El Helou, Daniel Sage, Thierry Laroche, Arne Seitz, Sabine Süsstrunk
To study JDSR on microscopy data, we propose such a novel JDSR dataset, Widefield2SIM (W2S), acquired using a conventional fluorescence widefield and SIM imaging.
1 code implementation • 7 Mar 2020 • Majed El Helou, Frederike Dümbgen, Sabine Süsstrunk
We propose to regularize this representation in the last feature layer before classification layers.
1 code implementation • 18 Dec 2019 • Siavash Bigdeli, David Honzátko, Sabine Süsstrunk, L. Andrea Dunbar
Plug-and-play denoisers can be used to perform generic image restoration tasks independent of the degradation type.
1 code implementation • 10 Dec 2019 • Chen Liu, Mathieu Salzmann, Sabine Süsstrunk
Training certifiable neural networks enables one to obtain models with robustness guarantees against adversarial attacks.
2 code implementations • 5 Jul 2019 • Majed El Helou, Sabine Süsstrunk
Blind and universal image denoising consists of using a unique model that denoises images with any level of noise.
Ranked #1 on
Grayscale Image Denoising
on BSD68 sigma40
1 code implementation • 20 May 2019 • Xiaoyan Zou, Ruofan Zhou, Majed El Helou, Sabine Süsstrunk
In this paper, we incorporate knowledge of the shadow's physical properties, in the form of shadow detection masks, into a correlation-based tracking algorithm.
1 code implementation • 9 May 2019 • Fayez Lahoud, Sabine Süsstrunk
Our method generates a single image containing features from multiple sources.
no code implementations • 2 Feb 2019 • Fayez Lahoud, Radhakrishna Achanta, Pablo Márquez-Neila, Sabine Süsstrunk
To obtain similar binary networks, existing methods rely on the sign activation function.
no code implementations • ICLR 2019 • Edo Collins, Siavash Arjomand Bigdeli, Sabine Süsstrunk
We propose a new notion of `non-linearity' of a network layer with respect to an input batch that is based on its proximity to a linear system, which is reflected in the non-negative rank of the activation matrix.
1 code implementation • 11 Sep 2018 • Majed El Helou, Frederike Dümbgen, Radhakrishna Achanta, Sabine Süsstrunk
Image optimization problems encompass many applications such as spectral fusion, deblurring, deconvolution, dehazing, matting, reflection removal and image interpolation, among others.
3 code implementations • ECCV 2018 • Edo Collins, Radhakrishna Achanta, Sabine Süsstrunk
We propose Deep Feature Factorization (DFF), a method capable of localizing similar semantic concepts within an image or a set of images.
Ranked #6 on
Unsupervised Human Pose Estimation
on Tai-Chi-HD
Unsupervised Facial Landmark Detection
Unsupervised Human Pose Estimation
+1
no code implementations • 29 May 2018 • Daniel Heydecker, Georg Maierhofer, Angelica I. Aviles-Rivero, Qingnan Fan, Dong-Dong Chen, Carola-Bibiane Schönlieb, Sabine Süsstrunk
Removing reflection artefacts from a single image is a problem of both theoretical and practical interest, which still presents challenges because of the massively ill-posed nature of the problem.
1 code implementation • 19 Feb 2018 • Ruofan Zhou, Radhakrishna Achanta, Sabine Süsstrunk
By training on high-quality samples, our deep residual demosaicing and super-resolution network is able to recover high-quality super-resolved images from low-resolution Bayer mosaics in a single step without producing the artifacts common to such processing when the two operations are done separately.
no code implementations • 27 Nov 2016 • Radhakrishna Achanta, Pablo Márquez-Neila, Pascal Fua, Sabine Süsstrunk
Since information is a natural way of measuring image complexity, our proposed algorithm leads to image segments that are smaller and denser in areas of high complexity and larger in homogeneous regions, thus simplifying the image while preserving its details.
no code implementations • 11 Nov 2015 • Ksenia Konyushkova, Nikolaos Arvanitopoulos, Zhargalma Dandarova Robert, Pierre-Yves Brandt, Sabine Süsstrunk
This paper introduces a novel approach to data analysis designed for the needs of specialists in psychology of religion.
no code implementations • 24 Jun 2014 • Neda Salamati, Diane Larlus, Gabriela Csurka, Sabine Süsstrunk
Based on a state-of-the-art segmentation framework and a novel manually segmented image database (both indoor and outdoor scenes) that contain 4-channel images (RGB+NIR), we study how to best incorporate the specific characteristics of the NIR response.