no code implementations • ECCV 2020 • Simon Jenni, Givi Meishvili, Paolo Favaro
Our representations can be learned from data without human annotation and provide a substantial boost to the training of neural networks on small labeled data sets for tasks such as action recognition, which require to accurately distinguish the motion of objects.
no code implementations • 15 Apr 2024 • Hamadi Chihaoui, Paolo Favaro
We introduce a novel approach to single image denoising based on the Blind Spot Denoising principle, which we call MAsked and SHuffled Blind Spot Denoising (MASH).
1 code implementation • 4 Apr 2024 • Alp Eren Sari, Francesco Locatello, Paolo Favaro
We present two practical improvement techniques for unsupervised segmentation learning.
no code implementations • 21 Mar 2024 • Aram Davtyan, Sepehr Sameni, Björn Ommer, Paolo Favaro
We call our model CAGE for visual Composition and Animation for video GEneration.
no code implementations • 29 Feb 2024 • Xiaohan Fei, Chethan Parameshwara, Jiawei Mo, Xiaolong Li, Ashwin Swaminathan, Cj Taylor, Paolo Favaro, Stefano Soatto
However, the SDS method is also the source of several artifacts, such as the Janus problem, the misalignment between the text prompt and the generated 3D model, and 3D model inaccuracies.
no code implementations • 7 Dec 2023 • Llukman Cerkezi, Aram Davtyan, Sepehr Sameni, Paolo Favaro
The growing interest in novel view synthesis, driven by Neural Radiance Field (NeRF) models, is hindered by scalability issues due to their reliance on precisely annotated multi-view images.
no code implementations • 3 Nov 2023 • Abdelhak Lemkhenter, Manchen Wang, Luca Zancato, Gurumurthy Swaminathan, Paolo Favaro, Davide Modolo
We show that SemiGPC improves performance when paired with different Semi-Supervised methods such as FixMatch, ReMixMatch, SimMatch and FreeMatch and different pre-training strategies including MSN and Dino.
no code implementations • 29 Sep 2023 • Zhuoran Yu, Manchen Wang, Yanbei Chen, Paolo Favaro, Davide Modolo
First, we introduce a denoising scheme to generate reliable pseudo-heatmaps as targets for learning from unlabeled data.
1 code implementation • 6 Sep 2023 • Llukman Cerkezi, Paolo Favaro
This sampling scheme relies on the mesh representation to ensure also that samples are well-distributed along its normals.
no code implementations • 28 Jun 2023 • Zhenlin Xu, Yi Zhu, Tiffany Deng, Abhay Mittal, Yanbei Chen, Manchen Wang, Paolo Favaro, Joseph Tighe, Davide Modolo
This paper introduces innovative benchmarks to evaluate Vision-Language Models (VLMs) in real-world zero-shot recognition tasks, focusing on the granularity and specificity of prompting text.
1 code implementation • 10 Jun 2023 • Josué Page Vizcaíno, Panagiotis Symvoulidis, Zeguan Wang, Jonas Jelten, Paolo Favaro, Edward S. Boyden, Tobias Lasser
Real-time 3D fluorescence microscopy is crucial for the spatiotemporal analysis of live organisms, such as neural activity monitoring.
no code implementations • CVPR 2023 • Yanbei Chen, Manchen Wang, Abhay Mittal, Zhenlin Xu, Paolo Favaro, Joseph Tighe, Davide Modolo
Our results show that ScaleDet achieves compelling strong model performance with an mAP of 50. 7 on LVIS, 58. 8 on COCO, 46. 8 on Objects365, 76. 2 on OpenImages, and 71. 8 on ODinW, surpassing state-of-the-art detectors with the same backbone.
Ranked #1 on Object Detection on OpenImages-v6 (using extra training data)
1 code implementation • 6 Jun 2023 • Aram Davtyan, Paolo Favaro
We propose a novel unsupervised method to autoregressively generate videos from a single frame and a sparse motion input.
no code implementations • CVPR 2023 • Achin Jain, Gurumurthy Swaminathan, Paolo Favaro, Hao Yang, Avinash Ravichandran, Hrayr Harutyunyan, Alessandro Achille, Onkar Dabeer, Bernt Schiele, Ashwin Swaminathan, Stefano Soatto
The PPL improves the performance estimation on average by 37% across 16 classification and 33% across 10 detection datasets, compared to the power law.
no code implementations • ICCV 2023 • Sepehr Sameni, Simon Jenni, Paolo Favaro
We propose Spatio-temporal Crop Aggregation for video representation LEarning (SCALE), a novel method that enjoys high scalability at both training and inference time.
no code implementations • ICCV 2023 • Aram Davtyan, Sepehr Sameni, Paolo Favaro
We call our model Random frame conditioned flow Integration for VidEo pRediction, or, in short, RIVER.
1 code implementation • 14 Oct 2022 • Adam Bielski, Paolo Favaro
We introduce MOVE, a novel method to segment objects without any form of supervision.
Ranked #2 on Unsupervised Saliency Detection on DUTS
no code implementations • 19 Sep 2022 • Luigi Fiorillo, Giuliana Monachino, Julia van der Meer, Marco Pesce, Jan D. Warncke, Markus H. Schmidt, Claudio L. A. Bassetti, Athina Tzovara, Paolo Favaro, Francesca D. Faraci
Specifically, we demonstrate that U-Sleep, a state-of-the-art sleep scoring algorithm, can be strong enough to solve the scoring task even using clinically non-recommended or non-conventional derivations, and with no need to exploit information about the chronological age of the subjects.
1 code implementation • 11 Aug 2022 • Zhaowei Cai, Avinash Ravichandran, Paolo Favaro, Manchen Wang, Davide Modolo, Rahul Bhotika, Zhuowen Tu, Stefano Soatto
We study semi-supervised learning (SSL) for vision transformers (ViT), an under-explored topic despite the wide adoption of the ViT architectures to different tasks.
no code implementations • 27 Jul 2022 • Abdelhak Lemkhenter, Paolo Favaro
In this work we introduce a novel meta-learning method for sleep scoring based on self-supervised learning.
no code implementations • 5 Jul 2022 • Luigi Fiorillo, Davide Pedroncelli, Valentina Agostini, Paolo Favaro, Francesca Dalia Faraci
Results: The performance of the models improves on all the databases when we train the models with our LSSC.
no code implementations • 13 Apr 2022 • Aram Davtyan, Paolo Favaro
We present GLASS, a method for Global and Local Action-driven Sequence Synthesis.
1 code implementation • 10 Apr 2022 • Sepehr Sameni, Simon Jenni, Paolo Favaro
We represent object parts with image tokens and train a ViT to detect which token has been combined with an incorrect positional embedding.
Ranked #91 on Image Classification on ObjectNet (using extra training data)
no code implementations • 14 Dec 2021 • Givi Meishvili, Attila Szabó, Simon Jenni, Paolo Favaro
Our method handles the complexity of face blur by implicitly learning the geometry and motion of faces through the joint training on three large datasets: FFHQ and 300VW, which are publicly available, and a new Bern Multi-View Face Dataset (BMFD) that we built.
no code implementations • 24 Aug 2021 • Luigi Fiorillo, Paolo Favaro, Francesca Dalia Faraci
We exploit, for the first time in sleep scoring, the Monte Carlo dropout technique to enhance the performance of the architecture and to also detect the uncertain instances.
no code implementations • 13 Jul 2021 • Abdelhak Lemkhenter, Adam Bielski, Alp Eren Sari, Paolo Favaro
We show a boost in the quality of generated samples with respect to FID from 10% to 40% compared to the baseline.
no code implementations • 7 Jul 2021 • Aram Davtyan, Sepehr Sameni, Llukman Cerkezi, Givi Meishvilli, Adam Bielski, Paolo Favaro
Moreover, we show that the Kalman Filter dynamical model for the evolution of the unknown parameters can be used to capture the gradient dynamics of advanced methods such as Momentum and Adam.
no code implementations • 18 Jun 2021 • Tomoki Watanabe, Paolo Favaro
However, because the classifier might still make mistakes, especially at the beginning of the training, we also refine the labels through self-attention, by using the labeling of real data samples only when the classifier outputs a high classification probability score.
no code implementations • 2 Apr 2021 • Adrian Wälchli, Paolo Favaro
In the case of synthetic data, the ground truth provides an exact and explicit description of what optical flow to assign to a given scene.
1 code implementation • ICCV 2021 • Ajinkya Tejankar, Soroush Abbasi Koohpayegani, Vipin Pillai, Paolo Favaro, Hamed Pirsiavash
Hence, we introduce a self supervised learning algorithm where we use a soft similarity for the negative images rather than a binary distinction between positive and negative pairs.
no code implementations • 13 Oct 2020 • Simon Jenni, Paolo Favaro
Current state-of-the-art methods cast monocular 3D human pose estimation as a learning problem by training neural networks on large data sets of images and corresponding skeleton poses.
1 code implementation • 16 Sep 2020 • Abdelhak Lemkhenter, Paolo Favaro
Various hand-crafted features representations of bio-signals rely primarily on the amplitude or power of the signal in specific frequency bands.
no code implementations • 21 Jul 2020 • Simon Jenni, Givi Meishvili, Paolo Favaro
Our representations can be learned from data without human annotation and provide a substantial boost to the training of neural networks on small labeled data sets for tasks such as action recognition, which require to accurately distinguish the motion of objects.
1 code implementation • 18 May 2020 • Josue Page, Paolo Favaro
We present a novel learning-based method to build a differentiable computational model of a real fluorescence microscope.
no code implementations • CVPR 2020 • Simon Jenni, Hailin Jin, Paolo Favaro
Based on this criterion, we introduce a novel image transformation that we call limited context inpainting (LCI).
1 code implementation • 24 Mar 2020 • Josue Page, Federico Saltarin, Yury Belyaev, Ruth Lyck, Paolo Favaro
To train our network, we built a data set of 362 light field images of mouse brain blood vessels and the corresponding aligned set of 3D confocal scans, which we use as ground truth.
no code implementations • 1 Oct 2019 • Attila Szabó, Givi Meishvili, Paolo Favaro
In this paper we present, to the best of our knowledge, the first method to learn a generative model of 3D shapes from natural images in a fully unsupervised way.
no code implementations • CVPR 2020 • Givi Meishvili, Simon Jenni, Paolo Favaro
To combine the aural and visual modalities, we propose a method to first build the latent representations of a face from the lone audio track and then from the lone low-resolution image.
no code implementations • CVPR 2019 • Simon Jenni, Paolo Favaro
We notice that the distributions of real and generated data should match even when they undergo the same filtering.
1 code implementation • NeurIPS 2019 • Adam Bielski, Paolo Favaro
To force the generator to learn a representation where the foreground layer corresponds to an object, we perturb the output of the generative model by introducing a random shift of both the foreground image and mask relative to the background.
no code implementations • 15 Mar 2019 • Tiziano Portenier, Qiyang Hu, Paolo Favaro, Matthias Zwicker
In this work, we propose a novel system for smart copy-paste, enabling the synthesis of high-quality results given a masked source image content and a target image context as input.
1 code implementation • 5 Dec 2018 • Qiyang Hu, Adrian Wälchli, Tiziano Portenier, Matthias Zwicker, Paolo Favaro
Because items in an image can be animated in arbitrarily many different ways, we introduce as control signal a sequence of motion strokes.
no code implementations • 26 Nov 2018 • Attila Szabó, Paolo Favaro
To achieve realism, the generative model is trained adversarially against a discriminator that tries to distinguish between the output of the renderer and real images from the given data set.
1 code implementation • ECCV 2018 • Simon Jenni, Paolo Favaro
Our approach is based on the principles of cross-validation, where a validation set is used to limit the model overfitting.
no code implementations • ECCV 2018 • Attila Szabo, Qiyang Hu, Tiziano Portenier, Matthias Zwicker, Paolo Favaro
We study the problem of building models that can transfer selected attributes from one image to another without affecting the other attributes.
no code implementations • ECCV 2018 • Meiguang Jin, Stefan Roth, Paolo Favaro
We introduce a family of novel approaches to single-image blind deconvolution, ie , the problem of recovering a sharp image and a blur kernel from a single blurry input.
no code implementations • CVPR 2018 • Simon Jenni, Paolo Favaro
To generate images with artifacts, we pre-train a high-capacity autoencoder and then we use a damage and repair strategy: First, we freeze the autoencoder and damage the output of the encoder by randomly dropping its entries.
no code implementations • CVPR 2018 • Mehdi Noroozi, Ananth Vinjimoor, Paolo Favaro, Hamed Pirsiavash
We use this framework to design a novel self-supervised task, which achieves state-of-the-art performance on the common benchmarks in PASCAL VOC 2007, ILSVRC12 and Places by a significant margin.
no code implementations • 24 Apr 2018 • Tiziano Portenier, Qiyang Hu, Attila Szabó, Siavash Arjomand Bigdeli, Paolo Favaro, Matthias Zwicker
We present a novel system for sketch-based face image editing, enabling users to edit images intuitively by sketching a few strokes on a region of interest.
1 code implementation • CVPR 2018 • Meiguang Jin, Givi Meishvili, Paolo Favaro
We present a method to extract a video sequence from a single motion-blurred image.
no code implementations • 8 Mar 2018 • Grigorios G. Chrysos, Paolo Favaro, Stefanos Zafeiriou
Notwithstanding, a much less standing mode of variation is motion deblurring, which however presents substantial challenges in face analysis.
no code implementations • CVPR 2018 • Qiyang Hu, Attila Szabó, Tiziano Portenier, Matthias Zwicker, Paolo Favaro
We learn our representation without any labeling or knowledge of the data domain, using an autoencoder architecture with two novel training objectives: first, we propose an invariance objective to encourage that encoding of each attribute, and decoding of each chunk, are invariant to changes in other attributes and chunks, respectively; second, we include a classification objective, which ensures that each chunk corresponds to a consistently discernible attribute in the represented image, hence avoiding degenerate feature mappings where some chunks are completely ignored.
2 code implementations • ICLR 2018 • Attila Szabó, Qiyang Hu, Tiziano Portenier, Matthias Zwicker, Paolo Favaro
Such models could be used to encode features that can efficiently be used for classification and to transfer attributes between different images in image synthesis.
1 code implementation • NeurIPS 2017 • Siavash Arjomand Bigdeli, Meiguang Jin, Paolo Favaro, Matthias Zwicker
We show that the gradient of our prior corresponds to the mean-shift vector on the natural image distribution.
Ranked #77 on Image Super-Resolution on Set14 - 4x upscaling
no code implementations • 22 Aug 2017 • Paramanand Chandramouli, Mehdi Noroozi, Paolo Favaro
In this paper, we address the problem of reflection removal and deblurring from a single image captured by a plenoptic camera.
2 code implementations • ICCV 2017 • Mehdi Noroozi, Hamed Pirsiavash, Paolo Favaro
In this paper, we use two image transformations in the context of counting: scaling and tiling.
no code implementations • CVPR 2017 • Meiguang Jin, Stefan Roth, Paolo Favaro
We present a novel approach to noise-blind deblurring, the problem of deblurring an image with known blur, but unknown noise level.
no code implementations • 5 Jan 2017 • Mehdi Noroozi, Paramanand Chandramouli, Paolo Favaro
The task of image deblurring is a very ill-posed problem as both the image and the blur are unknown.
9 code implementations • 30 Mar 2016 • Mehdi Noroozi, Paolo Favaro
By following the principles of self-supervision, we build a convolutional neural network (CNN) that can be trained to solve Jigsaw puzzles as a pretext task, which requires no manual labeling, and then later repurposed to solve object classification and detection.
no code implementations • 30 Nov 2014 • Daniele Perrone, Paolo Favaro
Our analysis reveals the very reason why an algorithm based on total variation works.
no code implementations • 16 Aug 2014 • Paramanand Chandramouli, Paolo Favaro, Daniele Perrone
We address for the first time the issue of motion blur in light field images captured from plenoptic cameras.
no code implementations • CVPR 2014 • Daniele Perrone, Paolo Favaro
This then results in a procedure that eludes the no-blur solution, despite it being a global minimum of the original energy.
no code implementations • CVPR 2013 • Thoma Papadhimitri, Paolo Favaro
We show that in the perspective projection case the solution is unique.