Efficient fine-tuning of pre-trained convolutional neural network (CNN) models using local
data is essential for providing high-quality services to users using ubiquitous and resource …

Adversarial training enhances neural network robustness but suffers from a tendency to
overfit and increased generalization errors on clean data. This work introduces CLAT, an …

The work addresses the issue of interactive image segmentation, relevant to modern
computer vision applications. The aim of the work is to improve the resolution of interactive …

We present a framework that enables an object detector to self-enhance its accuracy while
preserving its efficiency. This framework is particularly useful in settings where a single …

В данной работе исследуется проблема интерактивной сегментации изображений,
актуальная для современных приложений компьютерного зрения. Цель работы …

Les adaptations de rang faible (LoRa) sont devenues un standard pour adapter des
modèles à un faible coût. Elles sont de plus en plus utilisées que ce soit en traitement du …

In this paper, we propose using a domain-incremental learning approach for coping with
different devices in acoustic scene classification. While the typical way to handle …

Event and object detection in video streams is an important task in many computer vision
applications, from building surveillance and traffic monitoring to autonomous driving and …