Zobrazeno 1 - 10
of 1 819
pro vyhledávání: '"Sebe P"'
Autor:
Garosi, Marco, Tedoldi, Riccardo, Boscaini, Davide, Mancini, Massimiliano, Sebe, Nicu, Poiesi, Fabio
Supervised 3D part segmentation models are tailored for a fixed set of objects and parts, limiting their transferability to open-set, real-world scenarios. Recent works have explored vision-language models (VLMs) as a promising alternative, using mul
Externí odkaz:
http://arxiv.org/abs/2412.04247
Bioinformatics workflows are essential for complex biological data analyses and are often described in scientific articles with source code in public repositories. Extracting detailed workflow information from articles can improve accessibility and r
Externí odkaz:
http://arxiv.org/abs/2411.19295
Autor:
Li, Yawei, Ren, Bin, Liang, Jingyun, Ranjan, Rakesh, Liu, Mengyuan, Sebe, Nicu, Yang, Ming-Hsuan, Benini, Luca
While vision transformers show promise in numerous image restoration (IR) tasks, the challenge remains in efficiently generalizing and scaling up a model for multiple IR tasks. To strike a balance between efficiency and model capacity for a generaliz
Externí odkaz:
http://arxiv.org/abs/2411.18588
In this paper, we present an innovative solution for the challenges of the virtual try-on task: our novel Hierarchical Cross-Attention Network (HCANet). HCANet is crafted with two primary stages: geometric matching and try-on, each playing a crucial
Externí odkaz:
http://arxiv.org/abs/2411.15542
Regular unsupervised domain adaptive person re-identification (ReID) focuses on adapting a model from a source domain to a fixed target domain. However, an adapted ReID model can hardly retain previously-acquired knowledge and generalize to unseen da
Externí odkaz:
http://arxiv.org/abs/2411.14695
Autor:
Eerlings, Gilles, Vanbrabant, Sebe, Liesenborgs, Jori, Ruiz, Gustavo Rovelo, Vanacken, Davy, Luyten, Kris
We present an approach, AI-Spectra, to leverage model multiplicity for interactive systems. Model multiplicity means using slightly different AI models yielding equally valid outcomes or predictions for the same task, thus relying on many simultaneou
Externí odkaz:
http://arxiv.org/abs/2411.10490
Current face anonymization techniques often depend on identity loss calculated by face recognition models, which can be inaccurate and unreliable. Additionally, many methods require supplementary data such as facial landmarks and masks to guide the s
Externí odkaz:
http://arxiv.org/abs/2411.00762
In this paper, we study a practical yet challenging task, On-the-fly Category Discovery (OCD), aiming to online discover the newly-coming stream data that belong to both known and unknown classes, by leveraging only known category knowledge contained
Externí odkaz:
http://arxiv.org/abs/2410.19213
Referring 3D Segmentation is a visual-language task that segments all points of the specified object from a 3D point cloud described by a sentence of query. Previous works perform a two-stage paradigm, first conducting language-agnostic instance segm
Externí odkaz:
http://arxiv.org/abs/2410.13294
Autor:
Nikolaev, Sergey A., Chshiev, Mairbek, Ibrahim, Fatima, Krishnia, Sachin, Sebe, Nicolas, George, Jean-Marie, Cros, Vincent, Jaffrès, Henri, Fert, Albert
Publikováno v:
Nano Lett. 24, 43, 13465-13472 (2024)
Recent experiments by S. Krishnia et al., Nano Lett. 23, 6785 (2023) reported an unprecedentedly large enhancement of torques upon inserting thin Al layer in Co/Pt heterostructure that suggested the presence of a Rashba-like interaction at the metall
Externí odkaz:
http://arxiv.org/abs/2410.09957