6 papers
Action-Guided Attention for Video Action Anticipation
Tsung-Ming Tai, Sofia Casarin, Andrea Pilzer +2
Anticipating future actions in videos is challenging, as the observed frames provide only evidence of past activities, requiring the inference of latent intentions to predict upcom…
L-SWAG: Layer-Sample Wise Activation with Gradients information for Zero-Shot NAS on Vision Transformers
Sofia Casarin, Sergio Escalera, Oswald Lanz
Training-free Neural Architecture Search (NAS) efficiently identifies high-performing neural networks using zero-cost (ZC) proxies. Unlike multi-shot and one-shot NAS approaches, Z…
GRASP-GCN: Graph-Shape Prioritization for Neural Architecture Search under Distribution Shifts
Sofia Casarin, Oswald Lanz, Sergio Escalera
Neural Architecture Search (NAS) methods have shown to output networks that largely outperform human-designed networks. However, conventional NAS methods have mostly tackled the si…
Fractals as Pre-training Datasets for Anomaly Detection and Localization
C. I. Ugwu, S. Casarin, O. Lanz
Anomaly detection is crucial in large-scale industrial manufacturing as it helps detect and localise defective parts. Pre-training feature extractors on large-scale datasets is a p…
Your Image is My Video: Reshaping the Receptive Field via Image-To-Video Differentiable AutoAugmentation and Fusion
Sofia Casarin, Cynthia I. Ugwu, Sergio Escalera +1
The landscape of deep learning research is moving towards innovative strategies to harness the true potential of data. Traditionally, emphasis has been on scaling model architectur…
Spatiotemporal Modeling Encounters 3D Medical Image Analysis: Slice-Shift UNet with Multi-View Fusion
C. I. Ugwu, S. Casarin, O. Lanz
As a fundamental part of computational healthcare, Computer Tomography (CT) and Magnetic Resonance Imaging (MRI) provide volumetric data, making the development of algorithms for 3…