Search Results for author: Fabio Pizzati

Found 15 papers, 8 papers with code

Model Merging and Safety Alignment: One Bad Model Spoils the Bunch

no code implementations20 Jun 2024 Hasan Abed Al Kader Hammoud, Umberto Michieli, Fabio Pizzati, Philip Torr, Adel Bibi, Bernard Ghanem, Mete Ozay

Our experiments illustrate the effectiveness of integrating alignment-related data during merging, resulting in models that excel in both domain expertise and alignment.

On Pretraining Data Diversity for Self-Supervised Learning

1 code implementation20 Mar 2024 Hasan Abed Al Kader Hammoud, Tuhin Das, Fabio Pizzati, Philip Torr, Adel Bibi, Bernard Ghanem

We explore the impact of training with more diverse datasets, characterized by the number of unique samples, on the performance of self-supervised learning (SSL) under a fixed computational budget.

Diversity Self-Supervised Learning

SynthCLIP: Are We Ready for a Fully Synthetic CLIP Training?

1 code implementation2 Feb 2024 Hasan Abed Al Kader Hammoud, Hani Itani, Fabio Pizzati, Philip Torr, Adel Bibi, Bernard Ghanem

We present SynthCLIP, a novel framework for training CLIP models with entirely synthetic text-image pairs, significantly departing from previous methods relying on real data.

Material Palette: Extraction of Materials from a Single Image

no code implementations CVPR 2024 Ivan Lopes, Fabio Pizzati, Raoul de Charette

In this paper, we propose a method to extract physically-based rendering (PBR) materials from a single real-world image.

Unsupervised Domain Adaptation

ManiFest: Manifold Deformation for Few-shot Image Translation

1 code implementation26 Nov 2021 Fabio Pizzati, Jean-François Lalonde, Raoul de Charette

To enforce feature consistency, our framework learns a style manifold between source and proxy anchor domains (assumed to be composed of large numbers of images).

Image-to-Image Translation Translation

Leveraging Local Domains for Image-to-Image Translation

no code implementations9 Sep 2021 Anthony Dell'Eva, Fabio Pizzati, Massimo Bertozzi, Raoul de Charette

Our comprehensive evaluation setting shows we are able to generate realistic translations, with minimal priors, and training only on a few images.

Image-to-Image Translation Transfer Learning +1

Physics-informed Guided Disentanglement in Generative Networks

1 code implementation29 Jul 2021 Fabio Pizzati, Pietro Cerri, Raoul de Charette

Image-to-image translation (i2i) networks suffer from entanglement effects in presence of physics-related phenomena in target domain (such as occlusions, fog, etc), lowering altogether the translation quality, controllability and variability.

Disentanglement Image-to-Image Translation +1

CoMoGAN: continuous model-guided image-to-image translation

2 code implementations CVPR 2021 Fabio Pizzati, Pietro Cerri, Raoul de Charette

CoMoGAN is a continuous GAN relying on the unsupervised reorganization of the target data on a functional manifold.

Image-to-Image Translation Position +1

Model-based occlusion disentanglement for image-to-image translation

no code implementations ECCV 2020 Fabio Pizzati, Pietro Cerri, Raoul de Charette

Image-to-image translation is affected by entanglement phenomena, which may occur in case of target data encompassing occlusions such as raindrops, dirt, etc.

Disentanglement Image-to-Image Translation +1

Enhanced free space detection in multiple lanes based on single CNN with scene identification

2 code implementations2 May 2019 Fabio Pizzati, Fernando García

Traditional algorithms usually estimate only the position of the lanes on the road, but an autonomous control system may also need to know if a lane marking can be crossed or not, and what portion of space inside the lane is free from obstacles, to make safer control decisions.

Autonomous Driving Lane Detection

Cannot find the paper you are looking for? You can Submit a new open access paper.